Construction and analysis of novel controllable expression vectors for Bacillus subtilis

Transcription

1 Construction and analysis of novel controllable expression vectors for Bacillus subtilis Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften -Dr. rer. nat.- der Fakultät für Biologie, Chemie und Geowissenschaften der Universität Bayreuth vorgelegt von Phan Thi Phuong Trang aus Vietnam Bayreuth 2007

2 Die vorliegende Arbeit wurde in der Zeit von Juni 2004 bis Juni 2007 am Lehrstuhl für Genetik der Universität Bayreuth unter Leitung von Prof. Dr. W. Schumann durchgeführt. Vollständiger Abdruck der von der Fakultät für Biologie, Chemie und Geowissenschaften der Universität Bayreuth zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften genehmigten Dissertation. Promotionsgesuch eingereicht am: 14. März 2007 Wissenschaftliches Kolloquium am: 20. Juni Gutachter: Prof. Dr. Wolfgang Schumann 2. Gutachter: Prof. Dr. B. Westermann

3 Acknowledgements First and foremost, I would like to express my sincere gratitude to Professor Wolfgang Schumann for his wholehearted guidance, support and encouragement throughout this study and during my stay in Bayreuth. I am also grateful to Dr. Thomas Wiegert for his sincere advice and excellent discussion in the course of this work, especially for his kind help during my stay in Germany. I think highly of all the members in Prof. Schumann s Laboratory, who make our Laboratory a great place to work; I thank them for all their help, discussion and encouragement throughout this work. Thanks are due to Karin Angermann, Petra Helies and Brigitte Gubitz for their valuable assistance and making a warm atmosphere in the Laboratory. Furthermore, I am greatly indebted to Christa Schumann, who, like my mother, always makes me happy with all her heart. I do want to thank all of my friends in Bayreuth, especially Pham Dinh Trong, for their constant assistance and encouragement during my stay in Germany. I cannot help thanking my family who spiritually supports and encourages me in my study. Thanks a million for their taking care of our small son during my being away from him. I want to say special thanks to Nguyen Duc Hoang, my wonderful husband, who always motivates, consoles and discusses with me whenever I have difficulty in my study. I would not have finished my work without Nguyen Minh Nhut, who is not only our lovely son but also our companion, who does give me a great motivation in my study and life. Finally, I would like to thank Bayerische Forschungsstifung for its financial support. Phan Thi Phuong Trang

4 Contents Zusammenfassung...1 Summary Introduction Controllable expression systems in B. subtilis Published controllable expression system IPTG-inducible expression system Chromosomal integration systems Plasmid-based systems Riboswitches-based expression systems General characteristics of riboswitches The glycine riboswitch The lysine riboswitch Purification of recombinant proteins synthesized in B. subtilis General protein purification strategy Purification of His-tagged proteins Purification of Strep-tagged proteins Features for overproduction of proteins in B. subtilis Elements of a strong σ A -dependent promoter in B. subtilis The core promoter The UP element Messenger RNA stablizing elements Strong ribosome binding site Aims of the doctoral thesis Materials and methods Bacterial strains, plasmids, oligonucleotides, antibiotics and media Bacterial strains Plasmids...19 i

5 2.1.3 Oligonucleotides Antibiotics Media Enzymes, antibodies, biochemicals, chemicals and kits Enzymes Antibodies Biochemicals and chemicals Kits General methods PCR and colony PCR Cloning Growth and collection of samples Northern blot experiments Isolation of total RNA from B. subtilis Electrophoresis of RNA and vacuum blot transfer to membranes Transcriptional labelling of RNA probes Cleaning of DIG-labelling RNA probes Hybridization of membrane-bound RNA with RNA probes Stripping of RNA probes SDS-PAGE, Western blot analysis and rapid purification of proteins Extraction of denatured total cell lysate from B. subtilis Measurement of protein concentrations Precipitation of proteins from the culture supernatant Protein electrophoresis using discontinuous SDS-PAGE Immunoblot analysis Purification of proteins with His-tag and Strep-tag Visualization and measurement of reporter gene expression Visualization of extracellular enzyme activity (α-amylase) on plates Observation of the strength of promoters on X-gal plates Measurement of the β-galactosidase activity β-galactosidase BgaB...37 ii

6 β-galactosidase LacZ Construction of plasmids and strains Construction of expression vectors based on the glycine riboswitch Construction of expression vectors based on the lysine riboswitch Construction of structurally stable plasmids Construction of the promoter-probe vector pht Construction of plasmids to identify elements of strong promoters Construction of plasmids allowing the analysis of 3 stabilizing elements Construction of plasmids to study mrna stabilizing elements Construction of the knockout strains to study glycine expression system Construction of lysc knockout strains to study lysine expression system Results Exploring glycine controllable expression systems The glycine degrading gcv operon is strongly induced after addition of L-glycine Demonstration of the transcription attenuation model in-vivo To analyse for enhanced stability of the gcv riboswitch RNA To analyse for processing of the full-length transcript Detection of the DNA sequence coding for the transcriptional terminator Expression of the reporter gene lacz fused to glycine riboswitch To determine the optimal L-glycine concentration The effect of culture medium on β-galactosidase activity Does the growth temperature influence the expression level? Analysis of mutant gcv promoters Fusion of the groe promoter to the riboswitch Synthesis of recombinant proteins from plasmid-based expression systems Synthesis of intracellular recombinant proteins Synthesis of extracellular recombinant proteins Exploring lysine controllable expression systems Transcriptional analysis of the lysc gene Analysis for enhanced stability of the riboswitch RNA Analysis for processing of the full-length transcript...65 iii

7 Deletion of the DNA sequence coding for the transcriptional terminator Removal of L-lysine from the growth medium The auto-inducible expression systems Construction of plasmids allowing detection and single-step purification Removal of a 117-bp direct repeat results in structural stability Expression levels of the new expression vectors Incorporation of the epitope tag c-myc Addition of the His- and Strep-tags Addition of the His-tag Addition of the Strep-tag Establishment of a simple method for identification and screening Construction of the promoter-probe vector pht The use of the promoter-probe vector pht06 for cloning and analysis Linearity of BgaB activity and IPTG concentration Observation and measurement of the promoter strength Elements of strong σ A -dependent promoters in B. subtilis P grac is a strong σ A -dependent promoter for B. subtilis Influence of UP elements on gene expression Influence of the +1 region on gene expression Influence of the core promoter on gene expression levels Influence of the -10 and -35 regions on the BgaB activity Influence of the -15 region on the BgaB activity in B. subtilis and E. coli Analysis of combinations of the -10, the -15 and the -35 regions Combinations of UP elements and core promoters Using stabilizing elements to enhance the protein expression level mrna terminal stem-loops The 5 stem-loop structure (laco stabilizing element) Influence of a strong RBS Influence of the spacer length between the laco stem-loop and the RBS Combinations of strong stabilizing elements with different promoters Discussion iv

8 4.1 Riboswitches in expression systems The glycine system The lysine system A promoter-probe plasmid for screening promoters in B. subtilis Improve the productivity of B. subtilis Strong promoters in B. subtilis Role of messenger RNA stabilizing elements in overproduction of proteins Outlook References List of abbreviations and symbols v

9 Zusammenfassung Das Gram-positive Bakterium Bacillus subtilis spielt eine wichtige Rolle im Bereich der Landwirtschaft, der Medizin und der Ernährung und bei der Produktion rekombinanter Proteine. Momentan werden etwa 60% aller kommerziell verfügbaren technischen Enzyme von Bacillus-Spezies produziert. Außerdem sind eine Vielzahl von Informationen über die Transkription, die Translation, die Protein-Faltung, die Sekretions-Mechanismen, die genetische Manipulation und die Fermentation im industriellen Maßstab verfügbar. Dem steht gegenüber, dass effiziente und preiswerte Expressions-Vektoren bislang fehlen. Um diese Lücke zu schließen, wurden ein Glycin-induzierbares und ein Lysin-autoinduzierbares Expressionssystem entwickelt. Ferner wurden IPTG-induzierbare Expressions-Vektoren konstruiert und analysiert, die Überexpression und Reinigung von Proteinen erlauben. Weiterhin wurde ein Promoter-Testvektor entwickelt, der die Analyse von sehr starken Promotoren sowie von mrna stabilisierenden Elementen erlaubt, um die Menge an Transkript und die mrna-stabilität zu erhöhen und damit eine höhere Produktion an rekombinanten Proteinen zu gewährleisten. Während der Entwicklung der Glycin- und Lysin-induzierbaren Vektoren wurde mittels Northern-Blot bestätigt, dass in beiden Fällen zunächst ein kurzes Transkript, genannt Riboswitch, synthetisiert wird, welches nach Zugabe von Glycin und nach Entfernen von Lysin in ein längeres Transkript umgewandelt wird. Um die Expressionsstärke nach Induktion zu quantifizieren, wurden die beiden Promotoren mit ihren Riboswitches an das lacz- Reportergen fusioniert. Im Falle des Glycin-Systems wurde der Promotor optimiert und die Produktion von HtpG, Pbp4* und α-amylase als Modellproteine analysiert. Diese Ergebnisse zeigten, dass der Glycin-Riboswitch erfolgreich für die regulierte Produktion von sowohl intra- als auch extrazellulären Proteinen und der Lysin-Riboswitch als autoinduzierbares System verwendet werden können. Im letzteren Fall beginnt die Produktion der rekombinanten Proteine, wenn die Lysin-Konzentration in der Zelle unter einen Schwellenwert gefallen ist. Außerdem wurden sechs verschiedene neuartige IPTG-induzierbare Expressions-Vektoren für B. subtilis konstruiert. Während der erste Vektor die intrazelluläre Produktion von rekombinanten Proteinen erlaubt (pht01), enthält der zweite ein starkes Sekretions-Signal (pht43). Der dritte Vektor erlaubt das Anfügen des c-myc Epitop-Tags (pht10), und die restlichen drei Vektoren das Anfügen der His- (pht08) und Strep-Reinigungs-Tags (pht09 1

10 und pht24). Die Anwendung aller sechs Vektoren wurde durch das Einklonieren geeigneter Reportergene und die nachfolgende Überproduktion ihrer Proteine gezeigt. Zum schnellen Screening starker Promotoren wurde ein neuer Promotor-Test Vektor entwickelt mit dem bgab-gen als Reportergen, welches ein Blau-Weiß Screening erlaubt. Dieser Vektor enthält den laco-operator upstream von der Ribosomen-Bindungsstelle (RBS) und downstream von einer DNA-Sequenz mit Erkennungssequenzen für verschiedene Restriktionsenzyme. Er enthält außerdem das Gen für den Lac-Repressor, der die IPTGkontrollierte Expression der Promotoren erlaubt und somit die Klonierung auch starker Promotoren in E. coli, sowie die regulierte Expression von bgab in B. subtilis ermöglicht. Insgesamt 85 verschiedene und P groe -modifizierte σ A -abhängige Promotoren wurden in pht06 kloniert and analysiert. Es zeigte sich, dass DNA-Sequenzen um den Transkriptionsstart, die -10-, die -15- und die -35-Region und upstream vom Promotor dessen Stärke beeinflussen. Um die Aktivität der verschiedenen Promotoren zu vergleichen, wurden die BgaB-Aktivitäten bestimmt und Northern-Blot-Experimente durchgeführt. Die Messungen einiger neuer Kombinationen von Core-Promotoren und UP-Elemente zeigten, dass die β-galaktosidase-aktivitäten jeweils bis zu 13-fach bzw. 43-fach gesteigert werden konnten im Vergleich mit dem bereits starken P grac -Promotor. Wurden beide Elemente kombiniert, dann wurde eine etwa 690-fache Aktivität, wieder bezogen auf den P grac - Promotor, gemessen, und die Synthese an BgaB-Protein erreichte bis zu 30% des gesamten zellulären Proteins. Die mrna-stabilisierenden Elemente wurden mit einem vergleichbaren experimentellen Ansatz untersucht. Zunächst wurden 17 verschiedene Stem-Loop-Strukturen am 3'-Ende der mrna analysiert. Keine dieser DNA-Sequenzen zeigte einen signifikanten Einfluss auf die Menge an synthetisiertem Protein. Dann wurden 5'-mRNA-stabilisierende Elemente analysiert, und zwar eine starke RBS, ein laco-kontrollierbares stabilisierendes Element (genannt CoSE) und die Spacer-Regionen zwischen RBS und CoSE. Diese Ergebnisse zeigten, dass das CoSE-Element zusammen mit einem geeigneten Spacer und einer starken RBS die Genexpression 9-fach steigern konnte, wieder bezogen auf den P grac -Promotor. Dies führtezn bis zu 26% an rekombinantem Protein und einer Halbwertszeit der mrna von mehr als 60 min. Eine Kombination von starken Promotoren und stabilisierenden Elementen führtezn bis zu 42% an rekombinantem Protein als Anteil am Gesamtprotein. 2

11 Summary The gram-positive bacterium Bacillus subtilis is well-known for its contributions to agricultural, medical, and food biotechnology and for the production of recombinant proteins. At present, about 60% of the commercially available technical enzymes are produced by Bacillus species. Furthermore, a large body of information concerning transcription, translation, protein folding and secretion mechanisms, genetic manipulation, and large-scale fermentation has been acquired. But so far, efficient and inexpensive expression vectors for B. subtilis are still missing. To fill this gap, a glycine-inducible expression system and a lysine-autoinducible one were explored and IPTG-inducible expression plasmids that allow overexpression and purification of proteins were constructed and analyzed. Furthermore, a technique with a useful promoter-probe plasmid to analyze strong promoters in B. subtilis was established, which allowed to study promoter and mrna stabilizing elements to enhance the transcript level and mrna stability leading to higher production of recombinant protein. During the development of the glycine-inducible and lysine-autoinducible expression plasmids, the presence of a small transcript termed riboswitch corresponding to the 5' UTR in the absence of L-glycine or presence of L-lysine and its conversion into the full-length transcript after addition of the L-glycine or removal of L-lysine was confirmed by Northern blot. Next, the promoter and downstream riboswitch was fused to the lacz reporter gene to measure glycine or lysine-dependent induction. The production potential for the glycine system was analyzed in detail, and the promoter strength improved by using HtpG, Pbp4* and α-amylase as model proteins. In summary, the glycine riboswitch can be used successfully for regulatable production of both intra- and extracellular proteins, and the lysine riboswitch can be applied as an auto-inducible expression system allowing production of recombinant proteins when the L-lysine concentration within the growth medium falls below a threshold value. Six commercially available novel plasmid-based IPTG-inducible expression plasmids for B. subtilis were constructed, too. While the first vector allows intracellular production of recombinant proteins (pht01), the second provides a strong secretion signal (pht43). The third vector allows addition of the c-myc epitope tag (pht10), and the remaining three vectors provide the purification tags His (pht08) and Strep (pht09 and pht24). The versatility of all six vectors was verified by insertion of appropriate reporter genes and by demonstrating high level production of their proteins. 3

12 To develop a simple technique for rapid screening of strong promoters in B. subtilis, a new promoter-probe plasmid (pht06) using the bgab-encoded β-galactosidase was constructed allowing blue/white screening. This promoter-probe plasmid contains the laco operator, upstream of the ribosome binding site (RBS) and downstream of a multiple cloning sites (MCS), and laci coding for the Lac repressor, which allowed promoters to be controlled by IPTG, facilitating strong promoters to be cloned in E. coli and regulated expression of bgab in B. subtilis. A total of 85 different synthetic and groe-modified σ A -dependent promoters were introduced into pht06 and analyzed. Sequences around the transcriptional start site, the -10 region, the -15 region, the -35 region and the upstream region turned out to influence the promoter strength. BgaB activities and Northern blot analyses were used to measure the activity of the different promoters. The measurements of some new combinations of core promoters and UP elements on gene expression revealed that the β-galactosidase activity expression levels could be increased up to 13-fold and the mrna levels up to 43-fold as compared to the strong P grac promoter. If both elements were combined, an activity roughly 690 times higher than that obtained with the P spac promoter were obtained, and synthesis of BgaB, under control of these promoters, could reach up to 30% of the total cellular protein. The mrna stabilizing elements were also analyzed by using a similar experimental approach. First, seven-teen different 3 -mrna terminal stem-loops have been investigated, which did not significantly influence neither the amount of protein produced nor the mrna stability. Second, the 5 -mrna stabilizing elements including a strong RBS, the laco Controllable Stabilizing Element (CoSE) and the spacer between the RBS and CoSE were examined. The results demonstrated that CoSE together with an appropriate spacer and a strong RBS could increase gene expression 9-fold as compared to the P grac promoter, reaching up to 26% of total cellular protein and a half-life of the mrna of more than 60 min. A combination of strong promoters and stabilizing elements showed that recombinant protein synthesis levels of up to 42% of the total cellular protein could be obtained. 4

13 1 Introduction Bacillus species have been major workhorses in industrial microbiology since a very long time [56, 136]. The development of strains and production strategies has been influenced by the application of molecular biology techniques to strain development. Bacillus species are attractive industrial microorganisms due to several reasons such as high growth rate, their capacity to secrete proteins into the medium and their GRAS (generally regarded as safe) status with the Food and Drug Administration for some species including B. subtilis, the beststudied Gram-positive bacterium today. Expression systems for the production of recombinant proteins produced intra- and extracellularly have been reviewed recently [97, 163]. Bacillus subtilis is generally considered to have a great industrial potential for production and secretion of proteins of clinical interest like interferon [113], insulin [109], pathogenic antigens [2], and toxins [154], or technical enzymes of great industrial interest like proteases [63], α-amylase [66], and lipases [63]. The major advantages of B. subtilis as compared to other host production systems are high-cell-density growth and secretion of synthesized proteins into the cultivation medium, which facilitates isolation and purification of recombinant proteins during downstream processing [14, 17, 88]. High-level production of recombinant proteins as a prerequisite prior to subsequent purification has become a standard technique. Important applications of recombinant proteins are: (i) immunization, (ii) biochemical studies, (iii) three-dimensional analysis of the protein, and (iv) biotechnological and therapeutic use. Production of recombinant proteins involves cloning of the appropriate gene into an expression vector under the control of an inducible promoter. The selection of a particular expression system requires a cost breakdown in terms of design, process and other economic considerations. 1.1 Controllable expression systems in B. subtilis Expression systems for recombinant proteins rely on inducible promoter systems. Based on the inducing signal, promoters can be grouped into three classes: Those of class I are activated by adding a chemical compound which acts as an inducer. Cells are grown first in the absence of the inducer to the mid-logarithmic growth phase followed by addition of the inducer, a chemical compound such as IPTG or xylose. Expression of the recombinant gene in the absence of inducer is prevented by a repressor protein which can be inactivated by the added inducer such as the LacI repressor by IPTG or the XylR repressor by xylose [55]. Promoters 5

14 of class II are activated by a temperature shift, either up or down. Cells are grown at o C and shifted for induction of the promoter system either to high (40-45 o C) or low temperature (20-15 o C) [103, 122]. And promoters of class III are auto-inducible that direct low levels of expression in the lag and log phase, and much higher levels in the stationary phase. Their induction relies on the intracellular concentration of a metabolite such as the promoter of apre encoding for subtilisin E [71]. If this metabolite is present in excess, it will prevent expression of those genes involved in its own synthesis. During growth, the metabolite will be consumed by the cell, and if its concentration falls beyond a threshold value, the structural genes involved in its synthesis will be induced. And this class also includes promoters belonging to σ B -dependent promoters such as of gsib, encoding for a general stress protein [94] Published controllable expression system In order to produce homologous or heterologous proteins, several systems for inducible gene expression in B. subtilis have been developed. The starch-inducible amylase promoter is frequently used for production of heterologous proteins in which the desired protein is fused to the α-amylase promoter and leader peptide, which efficiently drives secretion of the protein produced into the culture medium [2, 63]. Several prophage derived heat-inducible gene expression systems that show very tight control of gene expression have been described. However, the levels of expression upon maximum induction are relatively low compared to those of other inducible gene expression systems [29, 63, 88]. The series of plasmid-based expression vectors phcmcs has been constructed allowing stable intracellular expression of recombinant proteins in B. subtilis cells. These expression vectors are based on the recently described Escherichia coli - B. subtilis shuttle vector pmtlbs72 that uses the theta mode of replication [156]. Three different controllable promoters have been inserted into the shuttle vector: P gsib that can be induced by heat, acid shock, and by ethanol, and P xyla and P spac that respond to the addition of xylose and IPTG, respectively. All recombinant vectors exhibited full structural stability [104]. An inducible gene expression system based on the regulation machinery of E. coli Tn10-encoded tetracycline resistance has been shown to be functional in B. subtilis [42]. This system has been reported to generate 100-fold-increased expression upon induction with tetracycline; however, considerable basal levels of expression are observed. A more tightly regulated variant of this system has been developed, but it appeared to generate lower maximal levels of expression upon induction [42]. Furthermore, a well-known system is the xyla system, in which a gene of interest is fused to the xylose-inducible xyla promoter, 6

15 which is integrated at the amye locus of the B. subtilis chromosome, has been reported to generate very high transcription activity upon xylose induction, whereas the basal level of expression is low [10, 79]. Recently, the SURE system, a SUbtilin-Regulated Expression system for B. subtilis has allowed strict control of gene expression by addition of subtilin. In this system, the spark-dependent signal transduction is used to control P spas -driven gene expression. Several multicopy expression vectors carrying subtilin-responsive promoter elements, which facilitated both transcriptional and translational promoter-gene fusions, have been constructed [14]. Very recently, a maltose-inducible expression vector in B. subtilis has been developed and characterized. The vector permitted β-galactosidase expression at a high level (maximum activity, 8.16 U/ml) when induced and its expression was markedly repressed by glucose. This provided a potential expression system for cloned genes in B. subtilis [98]. Finally, the well-known E. coli lac repressor-based expression system has been functionally implemented in B. subtilis as follows [117, 173] IPTG-inducible expression system The well-known E. coli lac repressor-based expression system has been functionally implemented in B. subtilis using a two-plasmid system, which allowed isopropyl-β-dthiogalactopyranoside (IPTG)-controlled gene expression in the latter species. This system was reported to exhibit no expression in the absence of the inducer, while very high levels of expression (10 to 15% of the total cellular protein) were observed after IPTG induction [55, 85, 117] Chromosomal integration systems This control mechanism is used in an expression system that employs the hybrid P spac promoter, which is composed of the B. subtilis phage SPO-1 promoter and the E. coli lac operator, in which IPTG-mediated derepression leads to transcription activation and yields high levels of gene expression. These plasmids allow the insertion of any kind of genetic information into the B. subtilis chromosome. The amye locus, coding for a nonessential α-amylase, is used in most cases for ectopic integration [55, 161, 173] Plasmid-based systems Two plasmid-based expression vectors have been constructed in which one plasmid allows intracellular production of recombinant proteins while the other directs the proteins into the culture medium. Both vectors use the strong promoter, P grac, which is composed of the B. subtilis groe promoter preceding the groesl operon (codes for the essential heat shock 7

16 proteins GroES and GroEL) of B. subtilis fused to the E. coli lac operator allowing their induction by addition of IPTG. While the background level of expression of these expression cassettes was very low in the absence of the inducer, an induction factor of about 1300 was measured upon addition of IPTG [117]. We and others observed that the groe promoter of B. subtilis is a strong promoter most probably due to the presence of an UP element [57, 96]. Based on these observations, this promoter was used to study whether it could drive expression of recombinant genes. Since the groe promoter is constitutive and high-level production of many recombinant proteins can be deleterious to the cells, a regulatory element had to be added. The lac operator (laco) of E. coli K12 was chosen, which had already been successfully used in combination with different promoters such as P spac [173] Riboswitches-based expression systems Each cell must regulate the expression of hundreds of different genes in response to changing environmental or cellular conditions. The majority of these sophisticated genetic control factors are proteins, which monitor metabolites and other chemical cues by selectively binding to targets. It has been discovered that RNA can also form precise genetic switches and that these elements can control fundamental biochemical processes. Riboswitches are a type of natural genetic control element that uses untranslated sequence in the 5 region of mrnas to form a binding pocket for a metabolite that regulates expression of that gene. During the last years the great importance of RNA for regulating gene expression in all organisms has become obvious. Consequently, several recent approaches aim to utilize the outstanding chemical properties of RNA to develop artificial RNA regulators for conditional gene expression systems. A combination of rational design, in vitro selection and in vivo screening systems has been used to create a versatile set of RNA-based molecular switches. These tools rely on diverse mechanisms and exhibit activity in several organisms, so they have been developmed recently in the application of engineered riboswitches for gene regulation in vivo [6] General characteristics of riboswitches As mentioned, riboswitches are metabolite binding domains located within the 5' untranslated regions (UTR) of some mrnas which are involved in gene regulation [159, 166]. Allosteric rearrangement of mrna structures is mediated by metabolite binding resulting in modulation of gene expression, and a change in expression with increasing ligand concentration, ranging from between 7-fold and 1,200-fold has been observed [50, 91, 95, 167] (Fig. 1.1). 8

17 A 1 B aptamer domain U U U U U A U G Fig Model of gene regulation by a typical riboswitch. (A) When the cellular concentration of metabolite is too low to occupy the riboswitch binding site, the transcription is completed, the biosynthetic and/or transport proteins are expressed; (B) when the cellular concentration of metabolite is high, the metabolite binds to X the riboswitch and leads to formation of an intrinsic terminator, the metabolite expression platform UUUUU biosynthetic or transport protein is not produced [41]. Riboswitches are conceptually divided into two parts: the aptamer and the expression platform (Fig. 1.1B). The aptamer directly binds the metabolite, and undergoes structural changes in response. These structural changes affect the expression platform, which is the mechanism by which gene expression is regulated. Expression platforms typically turn off gene expression in response to the metabolite, but some turn it on [37, 86, 159]. In the past year, three newly confirmed riboswitch classes have been reported (Fig. 1.2) [148, 159]. The first of these, the regulation of transcription termination, is utilized by nearly every riboswitch class and typically involves metabolite-dependent formation of a terminator stem, which prevents transcription elongation and inhibits gene expression (Fig. 1.2A). Two exceptions are the adenine and glycine riboswitch, wherein metabolite binding prevents terminator stem formation and activates gene expression [91-93, 148]. Second, the regulation of translation initiation is less widely utilized and involves altering the accessibility of the SD sequence (Fig. 1.2B). In this case, metabolite binding masks the SD sequence within a secondary structure to prevent ribosome binding and thereby inhibit gene expression. Interestingly, riboswitches in Gram-negative bacteria seemingly prefer regulation of translation initiation, whereas Gram-positive bacteria favour transcription termination, a correlation that probably reflects the higher frequency of polycistronic genes in Gram-positive 9

18 bacteria [108, 167]. For example, the TPP sensing riboswitch can terminate transcription of downstream genes in Gram-positive bacteria, suppress translation initiation in Gram-negative bacteria [135]. A third expression platform that can be utilized by riboswitches to affect gene expression is the regulation of RNA processing events. A conceptually simplistic manifestation of this expression platform is represented by the GlcN6P riboswitch, for which ligand binding induces catalytic self-cleavage of the mrna and inhibition of gene expression (Fig. 1.2C) [168]. However, it seems unlikely that the aptamer and expression platform (ribozyme) are separable functionalities, as they are for other riboswitches. Interestingly, the discovery of TPP-dependent riboswitches in eukaryotic genes has unveiled other possibilities for riboswitch control of RNA processing [83, 152]. For instance, the presence of TPP aptamers within introns or 3 untranslated regions (UTRs) suggests that riboswitches might regulate splicing or 3 end formation, respectively [135, 148, 159]. Fig Mechanisms of riboswitch function. (A) Transcription termination induced by metabolite (M) binding to nascent RNA, as observed for a guanine riboswitch; (B) translation initiation modulated by metabolite-dependent sequestration of a SD sequence, as observed for a TPP riboswitch; (C) RNA processing regulated by metabolitedependent self-cleavage, as observed for a GlcN6P riboswitch [148] The glycine riboswitch It has been suggested that about 2% of the B. subtilis genes are regulated via riboswitches [93], and three riboswitches have been studied so far. One of these riboswitches precedes the lysc gene [24]. The second is the gcvt operon involved in the degradation of L-glycine if the concentration is high within the cell [5, 93]. Both of these riboswitches operate by opposite mechanisms. The third are the members of the GlcN6P class of riboswitch which are selfcleaving ribozymes; they are activated when they are bound with the sugar-phosphate compound [148]. 10

19 The tricistronic gcvt-gcvpa-gcvpb operon codes for enzymes involved in the degradation of L-glycine if its concentration is high within the cell. The gcvt operon will be transcribed when the L-glycine concentration within the cell is high, and the metabolite will bind to a tandem riboswitch. The glycine riboswitch consisting of two strikingly similar aptamers, connected by a short linker region present upstream of glycine catabolism and efflux genes in a wide variety of bacteria. The glycine riboswitch binds L-glycine to regulate three glycine metabolism genes by activation via inhibition of premature termination of transcription, to use L-glycine as an energy source (type of regulation as Fig. 1.2A) [5, 93, 148, 159] The lysine riboswitch The lysc gene of B. subtilis encodes the inphase overlapping genes for the α- and β-subunits of a lysine-responsive aspartokinase II [24]. The lysine riboswitch (also called L-box) binds L-lysine to regulate lysine biosynthesis, catabolism and transport. The lysc gene is induced when the L-lysine concentration is low within the cell and the metabolite-free riboswitch favors formation of an anti-terminator structure. If the concentration of L-lysine is high in the cell, transcription of the lysine operon is initiated but terminated after a transcript of about 270 nucleotides has been synthesized. This 5 region of the lysine transcript is not translated, but forms a complicated secondary structure which is stabilized by L-lysine. This in turn leads to the formation of a terminator structure which causes the RNAP to dissociate from the DNA template and to release the transcript into the cytoplasm. This regulatory principle has been designated as riboswitch-mediated control of gene expression (type of regulation as Fig. 1.2A) [51, 153]. 1.2 Purification of recombinant proteins synthesized in B. subtilis An important application of gene technology is the overproduction of different proteins that can be utilized as pharmaceutical agents, as antigens for the production of antibodies, or as tools for structural and functional analyses. To separate one single protein from a complex mixture of proteins, while maintaining biological function, one can maintain biological function by controlling the ph, the temperature and the ionic strength (salt concentration) [11]. Many different proteins, domains, or peptides can be fused with the target protein. The advantages of using fusion proteins to facilitate purification and detection of recombinant proteins are well-recognized. The most frequently used and versatile systems are: Arg-tag, calmodulin-binding peptide, cellulose-binding domain, DsbA, c-myc-tag, glutathione S-transferase, FLAG-tag, HAT-tag, His-tag, maltose-binding protein, NusA, S-tag, SBP-tag, 11

20 Strep-tag, and thioredoxin [155]. Vectors allowing production of His-tagged proteins in B. subtilis have been published [142] General protein purification strategy To purify the protein of interest (separate it from other proteins in a mixture) one can take advantage of its general and specific properties: its native surface charge using ion exchange chromatography, its unique shape and size using gel filtration column chromatography, and its biological activity using affinity chromatography. These steps are sometimes applied in succession: first ion exchange chromatography to separate other proteins that have a different charge from the protein of interest, next gel filtration to separate all other proteins with a different size/shape than the protein of interest, and finally affinity chromatography to separate, based on biological activity, which is usually highly specific for the protein/enzyme of interest. In some cases, an enzyme may be purified to homogeneity (completely purified) using affinity chromatography alone since it is so effective at separating a specific enzyme from all other proteins in a mixture [11, 155] Purification of His-tagged proteins This protein purification system is based on the remarkable selectivity of the unique Ni-NTA (nickel-nitrilotriacetic acid) resin for recombinant proteins carrying a small affinity tag consisting of 6 to 10 consecutive histidine residues, termed the His-tag. The high affinity of the Ni-NTA resins for His-tagged proteins or peptides is due to both the specificity of the interaction between histidine residues and immobilized nickel ions and to the strength with which these ions are held to the NTA resin. NTA has a tetradentate chelating group that occupies four of six sites in the nickel coordination sphere. The metal is bound much more tightly than to a tridentate chelator such as IDA (imidodiacetic acid), which means that nickel ions made the proteins be very strongly bound to the resin. This allows more stringent washing conditions, better separation, higher purity, and higher capacity without nickel leaching [97, 121, 155] Purification of Strep-tagged proteins The Strep-tag is a selected eight-amino acid peptide (sequence: WSHPQFEK) with high specificity and affinity towards streptavidin; the Strep-tag has been developed as an alternative tool [139, 140]. Its sequence was derived by selection from a genetic peptide library [139]. The Strep-tag was bound at the same surface pocket where biotin, the natural ligand of streptavidin, gets complexed [138]. 12

21 The Strep-tag can be genetically fused up- or downstream of the reading frame of any gene and expressed as fusion protein. The Strep-tag system can be used to purify functional Streptag proteins from any expression system including baculovirus, mammalian cells, yeast, and bacteria. Because of its small size, the Strep-tag generally does not interfere with the bioactivity of the fusion partner [67]. 1.3 Features for overproduction of proteins in B. subtilis Control of gene expression can occur at the transcriptional or/and translation level (Fig. 1.3). Furthermore, gene expression can be controlled at the level of degradation of their mrnas. Different levels of gene expression are the result of varying frequencies of transcription and translation initiations. General features for overproduction of proteins are a high transcription rate (with strong promoters), low mrna degradation rate (including mrna stabilizing elements) and high translation rate (with strong RBS) [70, 77]. Fig Transcription and translation in a prokaryotic cell. Transcription and translation are coupled; that is, translation begins while the mrna is still being synthesized [12] Elements of a strong σ A -dependent promoter in B. subtilis Many housekeeping genes expressed during vegetative growth contain a typical σ A -dependent promoter, which is characterized by a -35 TTGACA consensus sequence and the -10 TATAAT hexanucleotide core elements and sometimes an UP element, in which, several weakly conserved A and T residues are present upstream of the -35 region (-36 to -70). The two hexanucleotide core elements are usually separated by a 17-nucleotide spacer sequence, and transcription is initiated around five nucleotides downstream of the -10 box (Fig. 1.4). Changes in their distance and in bases (even one single base) within these sequences can cause subtle to drastic changes in promoter activity. Altogether, some 4,000 genes are part of 13

22 the σ A regulon of B. subtilis, although their relative expression may vary significantly depending primarily, but not exclusively, on the actual sequence of the -35 and -10 elements [57, 101, 119] The core promoter The core promoter is the area from the -35 region to the transcription start site, which contains the canonical hexameric -35 box (essential for RNAP holoenzyme binding) and -10 box (essential for transcription initiation after binding has occurred), centred 10 and 35 bp upstream of the transcription start site. Bacteria have a multisubunit RNA polymerase (RNAP) with a conserved subunit composition. The core enzyme is composed of β, β, ω and two α subunits. Association of a σ subunit with the core enzyme forms the holoenzyme and determines the specificity of promoter utilization. Most RNAP holoenzyme molecules present during logarithmical growth contain the σ A factor [89, 96, 101, 119]. Extended promoter UP element Core promoter Region -10 Region Start site -70 Rich A + T TTGACA 16-18bp TATAAT +1 Terminator elements Docking site for the αctd ( * ) Sigma factor subunit specific DNA sequence recognition Fig Elements of the housekeeping σ A -dependent promoter. (*) C-terminal domain of the RNAP α subunit The UP element The UP element, located immediately 5 to the -35 element, has a recognizable pattern of ATrich sequences. It enhances RNAP binding by complexing with the C-terminal domain of α subunits and stimulates transcription initiation [57, 129]. In most cases, however, transcription of weak promoters is enhanced by regulatory proteins that act by binding to cognate and specific DNA sequences located upstream of the promoter and stimulating one or more steps of transcription initiation [89]. 14

23 The upstream promoter regions (-36 to -80) of B. subtilis σ A -dependent promoter sequences are enriched for short A and T tracts, suggesting that UP elements may be common for σ A - dependent B. subtilis promoters [57, 96] Messenger RNA stablizing elements Currently, at least 15 RNases are known in E. coli and at least 10 in B. subtilis [28]. The rate of mrna decay is an important element in the control of gene expression (Fig. 1.5). Given the absence of 5 to 3 exoribonucleolytic activities in prokaryotes, both endoribonucleases and 3 to 5 exoribonucleases are involved in chemical decay of mrna. As the 3 to 5 exoribonucleolytic activities are readily inhibited by stem-loop structures which are usual at the 3 ends of bacterial messages, the rate of decay is primarily determined by the rate of the first endonucleolytic cleavage within the transcripts, after which the resulting mrna intermediates with a 5 -monophosphate end is created, to which 3 to 5 exoribonucleases have greater affinity than the 5 -triphosphate end of the initial transcribed product. Successive cleavage events result in mrna fragments with accessible 3 -ends, which are rapidly degraded by 3-5 -exoribonucleases to oligonucleotides [35, 124]. Final turnover of mrna oligonucleotides to mononucleotides is accomplished by oligoribonuclease [43] (Fig 1.6). During the steps of mrna decay, stable RNA structures pose formidable barriers to the 3' to 5' exonucleases [3, 146]. DNA DNA Gene expression ON mrna protein gene Transcription Translation mrna degradation Gene expression OFF mrna protein gene Transcription mrna degradation No translation X Fig Novel mechanism in control of gene expression. When the rate of mrna degradation is low, most mrna molecules are translated (gene expression is ON; upper panel). When the rate of mrna degradation is high, most mrna molecules are degraded without translation (gene expression is OFF; lower panel). While 3 -terminal stem-loop structures play an important role as 5 stabilizers, 5 -proximal secondary structures or events such as ribosome stalling, regulatory protein binding, and ribosome binding can act as 5 stabilizers [7, 8, 44, 45, 53, 54, 134, 145]. 15

24 A B C Fig Model for mrna decay in E. coli. (A) mrna decay is initiated by the binding of RNase E to the 5' terminus of the transcript, followed by cleavage at an internal site ( ); (B) a polycistronic transcript is cleaved in an intergenic region ( ) by RNase III. For some transcripts (C) degradation does not involve any endonucleolytic cleavages but is carried out primarily by exonucleolytic attack by enzymes such as PNPase or RNase II. RNase G (restriction site ) does not bind efficiently to 5' termini that contain a triphosphate so it is hypothesized that it primarily cleaves degradation products that have been generated by either RNase E or RNase III. Dotted lines indicate inefficient pathways, 5'-triphosphates are shown in black while 5'-monophosphates are shown in green. Oligoribonuclease is necessary to degrade short oligoribonucleotides (4-7 nt) that are resistant to both PNPase and RNase II. This model is based on data published in [110] Strong ribosome binding site Translation in bacteria is initiated by interaction of the 3 end of the 16S rrna, which is part of the small ribosomal subunit, with the Shine Dalgarno sequence also called RSB. In B. subtilis, the consensus RBS sequence is AAAGGAGG, which is separated from the start codon by an about 7-nucleotide spacer sequence. The most frequently used start codon is 16

25 ATG (78%), but TTG (13%), GTG (9%), and CTG (<1%) are also used as translation starts [127]. High-level expression is not only dependent upon a strong regulatable promoter and 5 and 3 mrna stabilizers, but also on an efficient RBS sequence. One important example for a 5 stabilizer has been described as part of the gsib transcript, where a strong RBS enhances the half-life of the original transcript [74]. And so far, it was found that the 3' end of the cry gene of Bacillus thuringiensis conferred increased stability on other mrnas in both E. coli and B. subtilis [169]. 1.4 Aims of the doctoral thesis The major objections of the present doctoral thesis are as follows: (i) (ii) (iii) (iv) Development of a glycine-inducible expression system. Development of an autoinducible expression system relying on the consumption of L-lysine within the cell. Construction of a strong IPTG-inducible expression system based on the strong groesl promoter. Further enhancement of gene expression by optimizing the UP element, the region of the transcriptional start site and the half-life of the mrna. 17

26 2 Materials and methods 2.1 Bacterial strains, plasmids, oligonucleotides, antibiotics and media Bacterial strains The bacteria strains used in the course of this work are listed in the Tab Tab Bacterial strains used in this work Name Description Reference DH10B (E. coli) XL1 Blue (E. coli) F-, mcra, Δ(mrr, hsdrms, mcrbc), ϕ80d (laczδm15, ΔlacX74), deor, reca1, arad139, Δ(ara, leu) 7697, galk, λ -, rpsl, enda1, nupg reca1 enda1 gyra96 thi-1 hsdr17 supe44 rela1 lac [F proab laci q ZΔM15 Tn10 (Tet R )] Bethesda Research laboratories Stratagene 1012 leua8 metb5 trpc2 hsdrm1 [132] AM amye ::cat (Cm R ) [99] PT02 AM01 gcv::neo (Cm R and Neo R ) * PT gcvt::neo (Neo R ) * PT17 AM01 amye::p groe -Rib-lacZ (Spec R ) * PT21 AM01 amye::p gcv(-10 consensus) -lacz (Spec R ) * PT22 AM01 amye::p gcv(-35 consensus) -lacz (Spec R ) * PT23 AM01 amye::p gcv(-10 and -35 consensus) -lacz (Spec R ) * PT40 AM01 Δ gcv::neo (Cm R and Neo R ) * PT41 AM01 Δ lysc::neo (Cm R and Neo R ) * PT42 AM01 amye::p lysc -lacz (Spec R ) * PT43 PT41 amye::p lysc -lacz (Neo R and Spec R ) * PT44 AM01 amye::p lysc -Δter-laZ (Spec R ) * PT45 PT41 amye::p lysc -Δter-laZ (Neo R and Spec R ) * PT46 AM01 amye::p gcv -lacz (Spec R ) * PT47 PT40 amye::p gcv -lacz (Neo R and Spec R ) * PT48 AM01 amye::p gcv -Δter -lacz (Spec R ) * PT49 PT40 amye::p gcv -Δter -lacz (Neo R and Spec R ) * Strains marked with an asterisk (*) were constructed during this work. 18

27 2.1.2 Plasmids The plasmids used during this work are listed in the Tab Tab Plasmids used during this work Name Description Reference pdg1728 contains the promoter-less lacz allowing integration at amye [52] pmutin-ydrb template for lacz reporter gene [75] pmutin-gfp+ template for gfp+ reporter gene [75] pbgab template for t 0 terminator of phage λ [99] px-bgab template for bgab reporter gene [79] phcmc01 pmtlbs72 with trpa transcriptional terminator, resistant to Cm (Cm R ) [104] pndh09 template for srta gene of L. monocytogenes [106] pndh33 template for P groe promoter [117] pndh37 template for amyq signal sequence [117] pbluescript IIKS lacz, f1 ori, Amp R, T7 and T3 promoter Stratagene pct105 pbr322 + cela, template for cela [30] pct208 pbr322 + celb, template for celb [30] pt02z pt05-lacz with wild-type promoter P lysc -lys riboswitch * pt05z phcmc01 with lacz reporter gene * pt05-lacz pt05z with t 0 terminator of phage λ * pt12 pdg1728 with wild-type promoter P gcv -gly riboswitch * pt13 pdg1728 with wild-type promoter P lysc -lys riboswitch * pt17 pdg1728 with wild-type promoter P groe -gly riboswitch * pt20 Contains neomycin cassette for knockout of gcv operon * pt21 pdg1728 with promoter P gcv(-10 consensus) - gly riboswitch * pt22 pdg1728 with promoter P gcv(-35 consensus) - gly riboswitch * pt23 pt24 pdg1728 with promoter P gcv(-10 and -35 consensus) - gly riboswitch pht01 containing bgab fused to promoter P gcv -gly riboswitch * * 19

28 pt25 pht01 containing bgab fused to consensus promoter P gcv - gly riboswitch * pt27-htpg pt27-pbpe pt27-amyq pht01 containing htpg fused to promoter P gcv -gly riboswitch pht01 containing pbpe fused to promoter P gcv -gly riboswitch pht43 containing amyq fused to promoter P gcv -gly riboswitch * * * pt28-htpg pht01 containing htpg fused to consensus promoter P gcv - gly riboswitch * pt28-pbpe pht01 containing pbpe fused to consensus promoter P gcv - gly riboswitch * pt28-amyq pht43 containing amyq fused to consensus promoter P gcv -gly riboswitch * pt30 pht01 with wild-type promoter P gcv -gly riboswitch * pt31 pht01 with consensus promoter P gcv -gly riboswitch * pt32 pht43 with wild-type promoter P gcv -gly riboswitch * pt33 pht43 with consensus promoter P gcv -gly riboswitch * pt37 Contains neomycin cassette for knockout whole gcv operon * pt39 Contains neomycin cassette for knockout of lysc operon * pt40 pdg1728 with promoter P gcv -Δter gly riboswitch * pt41 pdg1728 with promoter P lysc -Δter lys riboswitch * pht01 derivative of pndh33 with deletion of a 117-bp direct repeat * pht06 promoter-probe plasmid to identify and screen promoters * pht36 plasmid for investigate terminal stabilizing elements * pht43 derivative of pndh37 with the deletion of a 117-bp direct repeat * pht08 pht01 with 8x His tag at the N terminus * pht08-yhcs pht08 with yhcs gene of B. subtilis * pht08-srta pht08 with srta gene of L. monocytogenes * pht09 pht01 with Strep-tag at the N terminus * 20

29 pht09-gfp+ pht09 with gfp reporter gene * pht10 pht01 with c-myc epitope-tag at the N terminus * pht10-ywbn pht10 with ywbn gene of B. subtilis * pht24 pht01 with Strep-tag at the C terminus * pht24-gfp+ pht24 with gfp reporter gene * pht42 pht08 with Strep-tag at the C terminus * Plasmids for the construction of promoter elements Plasmid Name of promoters/ oligos required for hybridization Reference pht57 P57/ P57F & P57R * pht58 P58/ P58F & P58R * pht59 P59/ P59F & P59R * pht60 P60/ P60F & P60R * pht61 P61/ P61F & P61R * pht62 P62/ P62F & P62R * pht68 P68/ P68F & P68R * pht69 P69/ P69F & P69R * pht70 P70/ P70F & P70R * pht71 P71/ P71F & P71R * pht72 P72/ P72F & P72R * pht73 P73/ P73F & P73R * pht74 P74/ P74F & P74R * pht75 P75/ P75F & P75R * pht76 P76/ P76F & P76R * pht77 P77/ P77F & P77R * pht78 P78/ P78F & P78R * pht79 P79/ P79F & P79R * pht80 P80/ P80F & P80R * pht81 P81/ P81F & P81R * pht82 P82/ P82F & P82R * 21

30 pht83 P83/ P83F & P83R * pht84 P84/ P84F & P84R * pht85 P85/ P85F & P85R * pht86 P86/ P86F & P86R * pht87 P87/ P87F & P87R * pht88 P88/ P88F & P88R * pht89 P89/ P89F & P89R * pht90/ pht95 P90/ P95/ P90F & P90R * pht91/ pht96 P91/ P96/ P91F & P91R * pht92/ pht97 P92/ P97/ P92F & P92R * pht93/ pht98/ P93/ P98/ P93F & P93R * pht94/ pht99 P94/ P99/ P94F & P94R * pht100 P100/ P100F & P100R * pht251 P251/ P251F & P251R * pht252 P252/ P252F & P252R * Intermediate PCR products required for generation of promoters Template/ Name of oligos PCR products P grac / S102R1 & ON76F P groe 01 P grac / S104R1 & ON76F P groe 02 P grac / S203R1 & ON76F P groe 03 P grac / S206R1 & ON76F P groe 04 P grac / S207R1 & ON76F P groe 05 P grac / S211R1 & ON76F P groe 06 Hybridization of S250F & S250R P groe 07 P grac / S228R & ON76F P groe 08 P grac / S229R & ON76F P groe 09 P lepa / S221F & S224R P lepa 224* Plasmids for the study of stabilizing elements Plasmid Templates/ oligos / name of stabilize element PCR products Reference pht102 P groe 01/ S102R2 & ON76F/ S102 * pht103 P groe 01/ S103R2 & ON76F/ S103 * pht104 P groe 02/ S104R2 & ON76F/ S104 * pht105 P groe 02/ S105R2 & ON76F/ S105 * 22

31 pht106 P grac / S106F & S106R/ S106 * pht107 P106/ S107R & ON76F/ S107 * pht108 P106/ S108R & ON76F/ S108 * pht109 P106/ S109R & ON76F/ S109 * pht201 P groe 01/ S201R2 & ON76F/ S201 * pht202 P groe 01/ S202R2 & ON76F/ S202 * pht203 P groe 03/ S202R2 & ON76F/ S203 * pht204 P groe 03/ S201R2 & ON76F/ S204 * pht205 P groe 03/ S104R2 & ON76F/ S205 * pht206 P groe 04/ S206R2 & ON76F/ S206 * pht207 P groe 05/ S105R2 & ON76F/ S207 * pht208 P groe 05/ S208R2 & ON76F/ S208 * pht209 P groe 02/ S208R2 & ON76F/ S209 * pht210 P groe 06/ S210R2 & ON76F/ S210 * pht211 P groe 06/ S211R2 & ON76F/ S211 * pht212 P groe 06/ S212R2 & ON76F/ S212 * pht213 P groe 06/ S213R2 & ON76F/ S213 * pht214 P groe 01/ S214R2 & ON76F/ S214 * pht215 P groe 01/ S215R2 & ON76F/ S215 * pht221 P lepa / S221R & S221F/ S221 * pht222 P spac 222*/ S223F & S212R2/ S222 * pht223 P groe 06/ S223F & S212R2/ S223 * pht224 P lepa 224*/ S221F & S212R2/ S224 * pht225 P groe 06/ S223F & S213R2/ S225 * pht228 P groe 08/ ON76F & S104R2/ S228 * pht229 P groe 09/ ON76F & S104R2/ S229 * pht250 P groe 07/ S250F & S212R2/ S250 * The name, description and references of the plasmids are given. Plasmids marked with an asterisk (*) were constructed during this work. 23

32 2.1.3 Oligonucleotides The oligonucleotides used during this work are listed in the Tab Tab Oligonucleotides used in the course of this work Name Sequence (5' to 3' ) Description ON01 ON02 CTAATACGACTCACTATAGGGAGAaaggacagagaaacacctcatgta ATATGAGCGAATGACAGCAAGG 3 end of glyriboswitch probe 5 end of glyriboswitch probe ON03 CTAATACGACTCACTATAGGGAGAagcattaatgacaagcagatag 3 end of gcvt probe ON04 GACCTGTATAAGGAATATGGAGGA 5 end of gcvt probe ON05 TAGATGGAGCTCAGAACGCCGTTATTTGACCTGT 5 end of gcvt gene ON06 CGCTGACCGCGGCTTCATCAATAAACGCAA 3 end of gcvt gene ON07 GGCCATCTCGAGGGCGCTTTACGTTTGATTATG 5 end of gcvpb gene ON08 GGCCATGGTACCGCCTCGTATCTGAGCACTG 3 end of gcvpb gene ON09 GGCCATGAATTCTTCAAACTCTGGAATTGCTAATG 5 end of P gcv ON10 ON11 GGCCATGGATCCTTCCTCCTTTATCAACGGCGCAGCT ggagattctttattataagaattgtccataacagcatgaaaatatg 3 end of P gcv - riboswitch recombinant primer for P groe - riboswitch ON12 TCGTTCGAATTCAGCTATTGTAACATAATCGGTACG 5 end of P groe ON13 GGAATTGTTATCCGCTCACAATTCCACAATTCTTATAATA 3 end of P groe ON14 ON15 ON16 ON17 ON18 ON19 GATGTAAGATATTGCTATAATATGTCCATAACAGCATGAAAA TATGAG GAGTATGTATTTGATGTAAGATATTGCTATAATATGTCCATAA CAGC GAGTTTGACATTGATGTAAGATATTGCTATAGTATGTCC GCATATAGTGATGATGGTAGGATATGAGTTTGACATTGATGT AAGATATTGCTATA GAGTTTGACATTGATGTAAGATATTGCTATAATATGTCCATAA CAGCATG GGCCATGAGCTCTTCAAACTCTGGAATTGCTAATG 5 end of -10 consensus 3 end of -10 consensus 5 end of -35 consensus 3 end of -35 consensus 5 end of -35 and -10 consensus 5 end of P gcv of B. subtilis 24

33 ON20 ON21 ON22 GGCCATTGATCAATGATTCAAAAACGAAAGCGGACAG GGCCATGGATCCTACGGCTGATGTTTTTGTAATCGG GGCCAGGATCCTTTCCCCTTTATCACACCTCATGTAAAATGAA GGTTCTC 5 end of signal sequence 3 end of signal sequence 3 end of glyriboswitch without terminator ON23 GGCCATGAGCTCCACTGTGACACAAGGGAAGC 5 end of yqhh gene ON24 GGCCATGGATCCAATGAATACAGAAATGATCTACGATG 3 end of yqhh gene ON25 ON26 GGCCATAGATCTCTAATTTCATAGTTAGATCGTGTTATATGG gctaatacgactcactatagggactctcattgcttatcaattaatcatcat 5 end of lysriboswitch probe 3 end of lysriboswitch probe ON27 CGCCAGAATTACAGATATCGACACTTC 5 end of lysc probe ON28 gctaatacgactcactatagggtatactcttcaagcaccgcaacgg 3 end of lysc probe ON29 GGCCATGAGCTCTGATCGGTGATCCGCTGG 5 end of uvrc gene ON30 GGCCATGGATCCTATCAGATCTTATTTAAAAGGACAACAT 3 end of uvrc gene ON31 GGCCATCTCGAGTCGCTTCACGATGCA 5 end of yslb gene ON32 GGCCATGGTACCCTCACCAACGTAAGCG 3 end of yslb gene ON33 GGCCATGAATTCACAAATTGCAAAAATAATGTTGTC 5 end of P lysc ON34 GGCCATGGATCCCATGTATTACCACCCTTTACATTTTG 3 end of P lysc ON35 GGCCATGGATCCTTCTCCCTTTCCTCTCATTGCTTATCAATTAA TCATCA 3 end of P lysc without terminator ON36 GGCCATTGATCAACAAATTGCAAAAATAATGTTGTC 5 end of P lysc ON37 GGCCATTCTAGACATGTATTACCACCCTTTACATTTTG 3 end of P lysc ON38 GCAGGATCCAAGGAGGAATCTAGAATGGAAGTTACTGACGTA AGATTACG 5 end of lacz gene ON39 GGCCATACTAGTTTATTTTTGACACCAGACCAACTGG 3 end of lacz gene ON40 ON41 GGCCATGCTAGCGATCTCTGCAGTCGCGATGAT GGCCATGGTACCGGGCAACGTTCTTGCCA 5 end of t 0 terminator 3 end of t 0 terminator ON42 GCCATCTCGAGGGTAACTAGCCTCGCCGATCC 5' amp-cole1 ON43 GCCATCTTAAGCATGCGTATTGGGCGCTCTTCCG 3' amp-cole1 25

34 ON44 GGCCATGGATCCATGAATGTGTTATCCTC 5' end of bgab ON45 GGCCATGACGTCCTAAACCTTCCCGGCTTCATCA 3' end of bgab ON46 GGCCATGGATCCATGGCGAAAAAAGAGTTTAAAGCAGAGTC 5' end of htpg ON47 GGCCATGACGTCTTACACCATGACCTTGCAAATATTGTTCG 3' end of htpg ON48 GGCCATGGATCCATGAAGCAGAATAAAAGAAAGCATC 5' end of pbpe ON49 GGCCATGACGTCTTACTACTTCGTACGGACCGCTTCT 3' end of pbpe ON50 GGCCATGGATCCATGATTCAAAAACGAAAGCGGACAG 5' end of amyq ON51 GGCCATGACGTCTTATTTCTGAACATAAATGGAGACG 3' end of amyq ON52 GGCCATAGATCTGCAAACACTGTGTCAGCGGCA 5' end of cela ON53 GGCCATGACGTCTTAATAAGGTAGGTGGGGTATGCTC 3' end of cela ON54 GGCCATGGATCCGAAGGGTCATATGCTGATTTGGCAG 5' end of celb ON55 GGCCATGACGTCTTATTTATACGGCAACTCACTTATGC 3' end of celb ON56 GATCTATGCGCGGAAGCCATCACCATCACCATCACCATCACG 5 end of 8xHis-tag ON57 GATCCGTGATGGTGATGGTGATGGTGATGGCTTCCGCGCATA 3 end of 8xHis-tag ON58 GATCTATGAATTGGAGCCATCCGCAATTTGAAAAAG 5 end of Strep-tag ON59 GATCCTTTTTCAAATTGCGGATGGCTCCAATTCATA 3 end of Strep-tag ON60 CGAACAAAAACTTATTAGCGAAGAAGATCTTTAATAACACGT 5 end of c-myc-tag ON61 GTTATTAAAGATCTTCTTCGCTAATAAGTTTTTGTTCGACGT 3 end of c-myc-tag ON62 CTGGAGCCATCCGCAATTTGAAAAATAAACGT 5 end of Strep-tag ON63 TTATTTTTCAAATTGCGGATGGCTCCAGACGT 3 end of Strep-tag ON64 GGCCAGATCTATGGGAGGCTTTAAATTGATCGATACG 5' end of yhcs gene ON65 GGCCATAGATCTAGAATGAAGAAAAGCCGCAGGCACT 3' end of yhcs gene ON66 GGCCATGGATCCATGATTAGTATATTTATTGCAGAAGA 5' end of desr gene ON67 GGCCATGACGTCTTATTTAAACCAGCCTTTTTCTTTTG 3' end of desr gene ON68 GGCCAGATCTATGAGTGGTCATGAAACAATCG 5' end of srta gene ON69 GGCGGATCCCGTTTTCGGCAGTTTCGTAGGGAAATATTTATTC TCTAGTT 3' end of srta gene ON70 GGCCATAGATCTATGAAGCCATCGAAAAAGGATGAAAAAG 3' end of ywbn gene ON71 GGCCATAGATCTTGATTCCAGCAAACGCTGGGC 5' end of ywbn gene ON72 GGCCATAGATCTATGGCTAGCAAAGGAGAAGAACT 5' end of gfp gene 26

35 ON73 GGCCATAGATCTTTTGTAGAGCTCATCCATGCCA 3' end of gfp gene ON74 GGCCATGACGTCTTATTTGTAGAGCTCATCCATGCCA 3' end of gfp gene ON75 GCTAATACGACTCACTATAGGGAACATTAATATCGCGCCAGT TAACATG 3 end of bgab probe ON76F TCGTTCGGTACCAGCTATTGTAACATAATCGGTACG 5 end of P groe ON77 GGCCATGGTACCAAAGGAGGTAAGGATCCATGAATGTGTTAT CCTC 5 end of bgab for pht02 ON78 ON79 ON80 GGCCATGACGTCCTAAACCTTCCCGGCTTCATCA GATCACTAGTTAACCGCGGAATTGTGAGCGGATAACAATTCC CATATAAAGGAGGAAG GATCCTTCCTCCTTTATATGGGAATTGTTATCCGCTCACAATTC CGCGGTTAACTAGT 3' end of bgab for pht02 MCS for pht06 MCS for pht06 P57F CTAGTTGACATTGGAAGGGAGATATGTTATTATAAGAATTGC 5 end of P57 P57R AATTCTTATAATAACATATCTCCCTTCCAATGTCAA 3 end of P57 P58F CTAGTTGAAATTGGAAGGGAGATATGTTATAATAAGAATTGC 5 end of P58 P58R AATTCTTATTATAACATATCTCCCTTCCAATTTCAA 3 end of P58 P59F P59R P60F P60R P61F P61R P62F P62R P63F P63R P64F P64R CTAGAAAATTTTTTAAAAAATCACTTGAAATTGGAAGGGAGA TTCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAAGTGATTT TTTAAAAAATTTT CTAGAAAATTTTTTATCTTATCACTTGAAATTGGAAGGGAGAT TCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAAGTGATAA GATAAAAAATTTT CTAGAAAATTTTTTATCTTATCAGTTGAAATTGGAAGGGAGAT TCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAACTGATAA GATAAAAAATTTT CTAGAAAATTTTTTATCTTACTACTTGAAATTGGAAGGGAGAT TCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAAGTAGTAA GATAAAAAATTTT CTAGAAAATTTTTTATCTTATCTCTTGAAATTGGAAGGGAGAT TCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAAGAGATAA GATAAAAAATTTT CTAGAAAATTTTTTAAAAAATCTCTTGAAATTGGAAGGGAGA TTCTTTATTATAAGAATTGC AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAAGAGATTT TTTAAAAAATTTT 5 end of P59 3 end of P59 5 end of P60 3 end of P60 5 end of P61 3 end of P61 5 end of P62 3 end of P62 5 end of P63 3 end of P63 5 end of P64 3 end of P64 P65F CTAGTAGACAAACTATCGTTTAACATGTTATACTATAATATGC 5 end of P65 27

36 P65R ATATTATAGTATAACATGTTAAACGATAGTTTGTCTA 3 end of P65 P66F CTAGTTGACACTTTATCTTCCATCTGGTATAATAAATAGAGC 5 end of P66 P66R TCTATTTATTATACCAGATGGAAGATAAAGTGTCAA 3 end of P66 P67F CTAGTTGACAAATATTATTCCATCTATTACAATAAATTCAGC 5 end of P67 P67R TGAATTTATTGTAATAGATGGAATAATATTTGTCAA 3 end of P67 P68F P68R P69F P69R CTAGAAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGA TATGTTATAATAAGAATTGC AATTCTTATTATAACATATCTCCCTTCCAATGTCAAGAGATTT TTTAAAAAATTTT CTAGAAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGA TTCTTTATAATAAGAATTGC AATTCTTATTATAAAGAATCTCCCTTCCAATGTCAAGAGATTT TTTAAAAAATTTT 5 end of P68 3 end of P68 5 end of P69 3 end of P69 P70F CTAGTTGAAATTGGAAGGGAGATTCTTTATTATAAGAATTGC 5 end of P70 P70R AATTCTTATAATAAAGAATCTCCCTTCCAATTTCAA 3 end of P70 P71F CTAGTTGACATTGGAAGGGAGATTCTTTATTATAAGAATTGC 5 end of P71 P71R AATTCTTATAATAAAGAATCTCCCTTCCAATGTCAA 3 end of P71 P72F CTAGTTGAAATTGGAAGGGAGATTCTTTATAATAAGAATTGC 5 end of P72 P72R AATTCTTATTATAAAGAATCTCCCTTCCAATTTCAA 3 end of P72 P73F CTAGTTGAAATTGGAAGGGAGATATGTTATTATAAGAATTGC 5 end of P73 P73R AATTCTTATAATAACATATCTCCCTTCCAATTTCAA 3 end of P73 P74F CTAGTTGAAATTGGAAGGGAGATTTGTTATTATAAGAATTGC 5 end of P74 P74R AATTCTTATAATAACAAATCTCCCTTCCAATTTCAA 3 end of P74 P75F CTAGTTGAAATTGGAAGGGAGAATGTTTATTATAAGAATTGC 5 end of P75 P75R AATTCTTATAATAAACATTCTCCCTTCCAATTTCAA 3 end of P75 P76F CTAGTTGAAATTGGAAGGGTTTATGTTTATTATAAGAATTGC 5 end of P76 P76R AATTCTTATAATAAACATAAACCCTTCCAATTTCAA 3 end of P76 P77F CTAGTTGAAATTGGAATATAGATTCTTTATTATAAGAATTGC 5 end of P77 P77R AATTCTTATAATAAAGAATCTATATTCCAATTTCAA 3 end of P77 28

37 P78F CTAGTTGACATTGGAAGGGAGATTCTTTATAATAAGAATTGC 5 end of P78 P78R AATTCTTATTATAAAGAATCTCCCTTCCAATGTCAA 3 end of P78 P79F CTAGTTGACATTGGAAGGGAGATATGTTATAATAAGAATTGC 5 end of P79 P79R AATTCTTATTATAACATATCTCCCTTCCAATGTCAA 3 end of P79 P80F CTAGTTGACATTGGAATATAGATATGTTATAATAAGAATTGC 5 end of P80 P80R AATTCTTATTATAACATATCTATATTCCAATGTCAA 3 end of P80 P81F CTAGTTGAAATTGGAATATTTTATGTTTATTATAAGAATTGC 5 end of P81 P81R AATTCTTATAATAAACATAAAATATTCCAATTTCAA 3 end of P81 P82F CTAGTTGACATTGGAATATAGATTCTTTATTATAAGAATTGC 5 end of P82 P82R AATTCTTATAATAAAGAATCTATATTCCAATGTCAA 3 end of P82 P83F CTAGTTGACATTGACAGGGAGATTCTTTATTATAAGAATTGC 5 end of P83 P83R AATTCTTATAATAAAGAATCTCCCTGTCAATGTCAA 3 end of P83 P84F CTAGTTGACATTGGAATATAGATATGTTATAATAAGAATAGC 5 end of P84 P84R TATTCTTATTATAACATATCTATATTCCAATGTCAA 3 end of P84 P85F CTAGTTGAATCTTTACAATCCTATTGATATAATCTAAGCTGC 5 end of P85 P85R AGCTTAGATTATATCAATAGGATTGTAAAGATTCAA 3 end of P85 P86F CTAGTTGACATTGGAATATAGATATGTTATAATAAGAATAGG GC 5 end of P86 P86R CCTATTCTTATTATAACATATCTATATTCCAATGTCAA 3 end of P86 P87F CTAGTTGACATTGGAATATTTTATGTTTATAATAAGAATTGC 5 end of P87 P87R AATTCTTATTATAAACATAAAATATTCCAATGTCAA 3 end of P87 P88F CTAGTTGACATTGGGTATAAGATTTGTTATAATAAGAATAGC 5 end of P88 P88R TATTCTTATTATAACAAATCTTATACCCAATGTCAA 3 end of P88 P89F CTAGTTGACATTGGTTTAAAGATATGTTATAATGGGTATAGC 5 end of P89 P89R TATACCCATTATAACATATCTTTAAACCAATGTCAA 3 end of P89 P90F CAAAAGAATGATGTAAGCGTGAAAAATTTTTTATCTTA 5 end of P90 29

38 P90R CTAGTAAGATAAAAAATTTTTCACGCTTACATCATTCTTTTGG TAC 3 end of P90 P91F CAAAAGAATGATAAAAGCGTGAAAAATTTTTTAAAAAA 5 end of P91 P91R CTAGTTTTTTAAAAAATTTTTCACGCTTTTATCATTCTTTTGGT AC 3 end of P91 P92F CTCACAAAAAAAGTGAGGATTTTTTTATTTTTGTA 5 end of P92 P92R CTAGTACAAAAATAAAAAAATCCTCACTTTTTTTGTGAGGTAC 3 end of P92 P93F CAACCCAGATATGATAGGGAACTTTTCTCTTTCTTGTTA 5 end of P93 P93R CTAGTAACAAGAAAGAGAAAAGTTCCCTATCATATCTGGGTT GGTAC 3 end of P93 P94F CTGTCAACATGAGAATTCTTATCATCAATTTTTGAAAA 5 end of P94 P94R P100F P100R CTAGTTTTCAAAAATTGATGATAAGAATTCTCATGTTGACAGG TAC CTAGAAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGA TATGTTATTATAAGAATTGC AATTCTTATAATAACATATCTCCCTTCCAATGTCAAGAGATTT TTTAAAAAATTTT 3 end of P94 5 end of P100 3 end of P100 P222F GGCCATGGTACCAGGCCTTACACAGCCCAGTCCAG 5 end of P222 P222R1 P251F P251R P252F P252R S102R2 S103R2 TGTGGAATTGTGATCCGCTCACAATCCACAATACACATTATGC CACACCTTGTAGAT CTAGAAAATTTTTTAAAAAATCTCTTGACAAATATTATTCCAT CTATGATTATAAATTCAGC TGAATTTATAATCATAGATGGAATAATATTTGTCAAGAGATTT TTTAAAAAATTTT CTAGAAAATTTTTTAAAAAATCTCTTGACACTTTACAATCCTA ATGATATTATCTAAGCTGC AGCTTAGATAATATCATTAGGATTGTAAAGTGTCAAGAGATTT TTTAAAAAATTTT GCCATGGATCCTTCCTCCTTTAATTGGTATCCGCTCACAATTC CACAA GGCAGGATCCTTCCTCCTTTAATTGAATTGTTATCCGCTCACA ATT 3 end of P222 5 end of P251 3 end of P251 5 end of P252 3 end of P252 3 end of S102 3 end of S103 S104R GGAATTGTTATCCGCTCACAATTCCCCTATTCTTATAATAAAG 3 end of S104 S104R2 S105R2 GGCCATGGATCCTTCCTCCTTTAATTGGGAATTGTTATCCGCT CACA GGCAGGATCCTTCCTCCTTTAATTGGGGAATTGTTATCCGCTC ACAATT 3 end of S104 3 end of S105 S106F GGCCATACTAGTGTACCAGCTATTGTAACATAATCGGTACG 5 end of S106 S106R GCTACCGCGGTATTCTTATAATAAAGAATCTCCCTTCC 3 end of S106 S107R GGCAGGATCCTTCCTCCTTTAATTGCGGAATTGTTATCCGCTC ACAATT 3 end of S107 30

39 S108R S109R S201R2 S202R2 S206R2 S208R2 S210R2 S211R2 S212R2 S213R2 S214R2 S215R2 S221R S223F S228R S229R GGCAGGATCCTTCCTCCTTTAATTCGCGGAATTGTTATCCGCT CACAATT GGCAGGATCCTTCCTCCTTTAATTCCGCGGAATTGTTATCCGC TCACAATT GGCAGGATCCTTCCTCCTTTAATTTGGAATTGTTATCCGCTCA CAATT GGCAGGATCCTTCCTCCTTTAATTTGTGGAATTGTTATCCGCT CACAATT GGCAGGATCCTTCCTCCTTTAATTGGGAATTGTGATCCGCTCA CAA GGCAGGATCCTTCCTCCTTTAGTTGGGGAATTGTTATCCGCTC ACAATT GGCAGGATCCTTCCTCCTTTAATTGGTTGAATTGATAGAATCT AGTTGTTGTGGAATTGTGATCCGCTCAC GGCAGGATCCTTCCTCCTTTAATTGGGAATTGTGATCCGCTCA CAA GGCAGGATCCTTCCTCCTTTAATTGGTGTTGGTTGTGGAATTG TGATCCGCTCACA GGCAGGATCCTTCCTCCTTTAATTGGTGTTGGTTGTTGTTGTG GAATTGTGATCCGCTCACA GGCAGGATCCTTCCTCCTTTAATTGGTGTTGGTTGTGGAATTG TTATCCGCTCACA GGCAGGATCCTTCCTCCTTTAATTGGTGTTGGTTGTTGTTGTG GAATTGTTATCCGCTCACA GGCAGGATCCTTCCTCCTTTAATTAAACGCAAAATACACTAGC TTAGAT CCGAGGTACCAAAATTTTTTAAAAAATCTCTTGACATTGGAAG GGAGATTCTTTATAATAAGAATTGTGG GGAATTGTTATCCGCTCACAATTCCGCTATTCTTATAATAAAG AATCTC GGAATTGTTATCCGCTCACAATTCCGCAATTCTTATAATAAAG AATCTC 3 end of S108 3 end of S109 3 end of S201 3 end of S202 3 end of S206 3 end of S208 3 end of S210 3 end of S211 3 end of S212 3 end of S213 3 end of S214 3 end of S215 3 end of S221 5 end of S223 3 end of S228 3 end of S229 ON, oligonucleotide; P, oligonucleotides for promoter elements; S, oligonucleotides for stabilizing elements; the DNA sequences recognized by restriction enzyme are underlined Antibiotics The antibiotics used in this work are listed in Tab Tab List of antibiotic solutions used in this work Concentration of Antibiotic stock solution (mg/ml) Dissolved in Final concentration (μg/ml) Ampicillin % ethanol 50 Chloramphenicol % ethanol 10 Rifampicin % methanol 100 Neomycin 10 water 10 Spectinomycin 100 water

40 2.1.5 Media LB medium (1 % (w/v) trypton, 0.5 % (w/v) yeast extract, 1 % (w/v) NaCl). Spizizen minimal medium (SMM) [149] supplemented with 50 µg/µl L-tryptophan, 50 µg/µl L- methionine and 50 µg/µl L-leucine. When necessary, the medium was supplemented with L-lysine at a final concentration of 300 µg/ml or L-glycine at a final concentration of 10 mm. Antibiotics (Table 2.4), 2 % (w/v) insoluble starch, 0.5 % CMC (w/v), 1.6 % X-gal and 0.01 or 0.1 or 1 mm IPTG were added when necessary. Agar was added to 1.5 % (w/v) to prepare plates. 2.2 Enzymes, antibodies, biochemicals, chemicals and kits Enzymes Roche: alkaline phosphatase, T7 RNA polymerase, DNase I. Merk: proteinase K. Sigma: RNase A, lysozyme. Fermentas: restriction enzymes, taq DNA polymerase, Deepvent DNA polymerase Antibodies All the antibodies used are given in Tab. 2.5 with their final dilutions. Tab Antibodies used in this work Name From organism Dilution Reference α-amyq Bacillus amiloliquefaciens 1 : V. Kontinen α-gfp Aequoria victoria 1 : 5000 Clontech α-htpg Bacillus subtilis : S. Schwab α-pbp * Bacillus subtilis : [176] α-his synthetic 1 : 5000 Qiagen α-strep synthetic 1 : 5000 IBA α-c-myc synthetic 1 : 5000 Sigma Biochemicals and chemicals Amersham: Amonium persulphate, hyperfilm ECL. Fermentas: DNA ladder, RNA ladder and protein ladder. 32

41 Pierce: Luminol substrate. Roche: blocking reagent, chemiluminescent substrate CPD-Star, protease inhibitor, RNAase inhibitor, Xgal, ONPG, DTT and IPTG. Roth: acetic acid, agar, agarose, aqua phenol, chloroform, diethylpyrocarbonate (DEPC), Ethidium bromide (Et.Br), isopropanol, L-lysine, L-glycine, L-Tryotophan, L-methionine, L- leucine, pepton, potassium acetate, potasium phosphate, polyacrylamide, sodium phosphate, sodium chloride, starch, MOPS, sodium dodecyl sulphate, TEMES, Tris, yeast extract Kits Epicentre: Fast-Link DNA Ligation Kit. Qiagen: PCR purification Kit, gel-extraction Kit, midi purification Kit and Ni-NTA Spin (50) Kit. IBA: Strep-Tactin Spin Column Kit. 2.3 General methods PCR and colony PCR The technique of PCR (Polymerase Chain Reaction) is used to produce a large number of copies of a DNA sequence, e.g., a gene, to provide a sufficient amount for cloning [131]. This became effective by the isolation of a thermostable DNA polymerase from Thermus aquaticus [130]. During the PCR, DNA is denatured at high temperature and specific oligonucleotide primers are annealed and elongated at lower temperature in a cyclic manner. Colony PCR is based on standard PCR using total DNA from colonies as template to allow the rapid detection of recombinant clones Cloning All the steps necessary for cloning were carried out as described by using standard methods [133]. Preparation of competent E. coli cells and transformation were carried out as standard heat shock transformation [68] or electroporation [36]; PCR for screening of plasmids and preparation of plasmid DNA by the alkaline lysis method with SDS have been described [13, 69]. The correct DNA sequence of all inserts into plasmids was verified by sequencing carried out by the SeqLab, and only plasmids with the correct DNA sequence were used in further experiments. 33

42 2.3.3 Growth and collection of samples During this work, B. subtilis strains were grown in LB medium or SMM with the appropriate antibiotic(s) when necessary in a waterbath shaker ( 200 rpm) at 37 o C. Overnight cultures in 5 ml medium in glass tubes were transferred partially to Erlenmeyer flasks containing medium to an OD 578 of When an OD 578 of 0.8 (in LB medium) or 0.2 (in SMM) was reached (set as t = 0), the culture was divided into two subcultures where one was further grown in the absence and the other in the presence of the inducer, 0.01 mm or 0.1 mm or 1 mm for IPTG or 10 mm for L-glycine or 300 µg/ml for L-lysine. Aliquots were removed and centrifuged, and either the pellet and/or the culture supernatant were collected. Further samples were taken at different points of time after induction as indicated in the experiments. Normally, a certain amount of cells was collected corresponding to 1.2 of OD 578 for β-galactosidase assays, or 2.5 of OD 578 for protein analyses and 10 of OD 578 for RNA analyses. 2.4 Northern blot experiments Northern blot analysis was performed as described [65, 128] Isolation of total RNA from B. subtilis B. subtilis cells were grown and induced as described in Especially, B. subtilis strains 1012, PT43, PT45, PT47, PT49 and B. subtilis 1012/pHT plasmids were induced by removal of L-lysine or 10 mm L-glycine or 0.1 mm IPTG and/or 100 µg/ml of rifampicin were added when necessary. The cells were then killed by addition of killing buffer (5 mm MgCl 2, 20 mm NaN 3, 20 mm Tris-HCl; ph7.5). Total RNA was extracted using the protocol for isolation of RNA from yeast with modifications [126]. The cell walls of the cells were digested by lysozyme (1 mg/ml) on ice for 5 min before extraction of the RNA. The samples were heated at 95 o C for 5 min before addition of phenol Electrophoresis of RNA and vacuum blot transfer to membranes RNA samples were separated on 1.2% agarose gels and the transfer occurred onto Nylonmembranes. The transfer was carried out with the help of the Vacuum-Blot-Annex (VacuGene TM X1, Pharmacia) Transcriptional labelling of RNA probes Pairs of primers ON44/ON75 were used to amplify an internal part of the bgab gene. The primer pairs ON01/ON02 and ON25/ON26 were used to generate the PCR product 34

43 corresponding to the lysc-riboswitch and gcv-riboswitch region and ON03/ON04 and ON27/ON28 to an internal part of lysc and gcv, respectively. These amplicons harbour a T7 promoter at the 3 end and were used as templates for in vitro transcription according to the instructions of the manufacture [128] Cleaning of DIG-labelling RNA probes When the DIG-labelled-antisense-RNA was used at the beginning for hybridization experiments, a very strong background was detected on the X-ray-film. The more RNA probe was used, the more the background was decreased. While this phenomenon has been observed, its reasons are not known. Therefore, the RNA probes were purified routinely before they were used in hybridization experiments. All the steps were carried as in the hybridization procedures, but the blank membranes were used Hybridization of membrane-bound RNA with RNA probes This experiment was carried out as described [128] Stripping of RNA probes This experiment was carried out as described [128]. 2.5 SDS-PAGE, Western blot analysis and rapid purification of proteins Extraction of denatured total cell lysate from B. subtilis For the extraction of denatured cell lysate from B. subtilis cells (2.5 of OD 578 ) prepared as in were resuspended in 100 μl of lysis buffer (15% (w/v) sucrose, 50 mm Tris/HCl; ph 7.2) containing 2.5 mg/ml lysozyme and incubated at 37 o C for 5 min. Then, 50 μl of 3x sample loading buffer (0.135 M Tris/HCl, 30% glycerol, 3% SDS, 0.03% bromophenol blue, 0.15 M DTT) were added to the suspension and frozen until they would be used. Before the samples were used, they had been heated for 5 min at 95 o C and 15 μl of each sample were used for SDS-PAGE Measurement of protein concentrations The method of Bradford was used for the measurement of the protein concentrations from cell extracts [81] Precipitation of proteins from the culture supernatant Protein from cultured supernatant was collected by the TCA method. One volume of 40% TCA was mixed with three volumes of culture supernatant, incubated on ice for 10 min and 35

44 centrifuged (12,000 rpm at 4 o C for 10 min). The pellet was then washed twice with ice-cold acetone and dried at room temperature. The pellet was dissolved in water and loading buffer for SDS-PAGE was added Protein electrophoresis using discontinuous SDS-PAGE The electrophoretic separation of proteins according to their molecular mass was performed as first described by Laemmli [84] Immunoblot analysis In order to immunochemically detect proteins using antibodies, the proteins were transferred, after their electrophoretic separation, onto a nitrocellulose membrane using electroblotting [158]. The electrophoretic transfer of proteins to nitrocellulose membranes was achieved by Semi-Dry-Blotting between graphite plate electrodes in a Fast-Blot apparatus (Biorad). The procedure for detection of labelled proteins followed the instruction of ECL Western blot (Amersham Biociences). To corroborate the versatilities of the expression vectors pt and pht for the overproduction of proteins, B. subtilis 1012 harbouring plasmids with the heat shock gene htpg [143] (pt27- htpg, pt28-htpg and pht01-htpg) and the gene pbpe coding for a penicillin binding protein (pt27-pbpe, pt28-pbpe and pht01-pbpe) were analysed by immunoblot. B. subtilis strain 1012 carrying one of these two vectors was grown either in LB medium or SMM [149] at 37 o C to the mid-exponential growth phase. Then, the cultures were divided into subcultures where one was further grown untreated while the others were induced with 10 mm L-glycine or 1 mm IPTG. These cells were further grown, and samples were collected as described in and prepared (see under 2.5.1). Equal amounts of proteins (0.025 of OD 578 ) were applied per lane, and the blots were probed with α-htpg or α-pbp4*. To confirm expression of α-amylase in the culture medium, cells of B. subtilis 1012 containing the plasmids pt27-amyq or pt28-amyq or pht43-amyq were grown as described in Aliquots were taken immediately before addition of 10 mm L-glycine or 1 mm IPTG (0 h) and 2 to 4 h after induction. Cells were pelleted by centrifugation and equal amounts of supernatant were analyzed, and the blots were probed with α-amyq. In addition, to confirm the presence of the 10-amino-acid epitope-tag (c-myc, pht10-ywpn and pht10-gfp) or the 8-amino-acid His (8xHis, pht08-yhcs and pht08-srta) and Strep 36

45 (pht09-gfp and pht24-gfp) purification tags, a Western blot was carried and the monoclonal antibodies α-c-myc, α-his or α-strep were used. All other cells were grown as described under 2.3.3; samples were applied on polyacrylamide gels and transferred to membranes for immunoblot detection as described in this section Purification of proteins with His-tag and Strep-tag In order to allow purification of recombinant proteins from bacterial lysates by single-step of affinity chromatography, the proteins were fused with two different purification tags, namely the His- and the Strep-tag. The purification procedure can be divided into three stages: preparation of the cell lysate, binding of the recombinant protein to the affinity column, washing, and elution of the recombinant protein. The method of single-step purification of His-tagged proteins is described in [121] and that of Strep-tagged proteins in [67]. 2.6 Visualization and measurement of reporter gene expression Visualization of extracellular enzyme activity (α-amylase) on plates Single colonies of the B. subtilis strains carrying plasmid pt27-amyq or pt28-amyq were grown for 24 h on LB or SMM plates with or without 10 mm L-glycine and 2% insoluble starch and stained with I2/KI solution [107]. Pictures were recorded using a digital camera Observation of the strength of promoters on X-gal plates Blue colonies can be seen on X-gal LB plates containing IPTG. However, to see the difference of the promoter strength, low amount of IPTG should be added. Especially, to screen for strong promoters, 0.01 mm IPTG was applied and the plates with B. subtilis strains could be observed between h for blue and white colonies. To compare B. subtilis strains carrying different plasmids, cells were transferred to new plates with various concentrations of IPTG (0, , 0.005, 0.01, 0.025, 0.05 and 0.1 mm) and incubated at 30 o C or 37 o C. Each sample was replicated twice. The grey values (density) of colonies were determined using QuantityOne programme (Biorad) and represents the strength of promoters; higher values mean stronger promoters Measurement of the β-galactosidase activity β-galactosidase BgaB Blue colonies from LB-Xgal plates were used for determination of β-galactosidase activity. B. subtilis strains 1012 carrying pt or pht with the bgab reporter gene were grown, and 37

46 samples collected as described under The activity was determined at 55 o C as described [99] with the exception that the BgaB activity was measured in a microtiterplate reader (VersaMax, Molecular Devices). One unit is defined as ΔA 420 *OD *min -1 and displayed as units/od 578 for all the results, in which one OD 578 is defined as the optical density of the samples used in the assay, A 420 is the absorbance of the samples measured by the microtiterplate reader and, min indicates the incubation time of the plate at 55 o C β-galactosidase LacZ B. subtilis strains PT17, PT21, PT22, PT23, PT42 to PT49 and 1012 carrying pt02z were grown and samples were collected according to section β-galactosidase activity were measured at 405 nm at 28 o C as described elsewhere [164]. One unit is defined as V max *OD , in which OD 578 is defined as the optical density of the samples used in the assay and V max is the maximum kinetic rate reported as miliod/minute. The data were displayed as units/od 578 for all the results. 2.7 Construction of plasmids and strains Construction of expression vectors based on the glycine riboswitch To allow integration of transcriptional fusions at the amye locus, the integration vector pdg1728 was used. This vector contains a promoter-less lacz gene sandwiched between amye-front and amye-back [52] and was used as a backbone for the construction of several plasmids. pt12 contains the wild-type promoter of the gcv operon and the complete riboswitch amplified by PCR using the oligos ON9 and ON10 and DNA of strain 1012 as template inserted into pdg1728. pt17 harbours the promoter of the groe operon (ON12/ON13; template pndh33) fused to the riboswitch (ON10/ON11; template DNA 1012); both PCR products were fused using recombinant PCR. pt21 carries the -10 consensus region of the gcv promoter and was obtained by using the mutagenic primer ON15 together with ON9 to amplify the promoter region and ON10 and ON14 to amplify the riboswitch followed by recombinant PCR to fuse both amplicons. pt22 contains the -35 consensus sequence and was constructed in a similar way using ON9 and the mutagenic primer ON17 to amplify the promoter region and ON10 and ON16 for the riboswitch. pt23 carries the complete consensus sequence obtained with the primer pairs ON9 and ON17 for the promoter and ON10 and ON18 for the riboswitch region. To study the influence of the transcriptional terminator on the regulation of gcv, the region corresponding to the right arm of the inverted repeat was removed using the primer pair ON9 and ON22 resulting in pt40. ON22 contains 38

47 the SD sequence of gcv to ensure translation of lacz. All amplicons were verified by DNA sequencing. The resulting plasmids pt12, pt17, pt21, pt22, pt23, pt40 were transformed into the B. subtilis strain AM01 selecting for spectinomycin-resistant cells and screening for chloramphenicol sensitivity, and strain PT46, PT17, PT21, PT22, PT23 and PT48 were kept for further studies. The transcriptional fusion from plasmids pt12 and pt40 were also transformed into strain PT40 yielding PT47 and PT49 [116]. The following constructs allow expression of recombinant genes from two plasmid vectors: pht01 carrying either the promoter-less bgab or htpg or pbpe gene and pht43 carry the promoter-less amyq gene including its own signal sequence. The wild-type gcv promoter together with the riboswitch (ON10/ON19; template pt12) was fused to bgab (pt24), htpg (pt27-htpg), pbpe (pt27-pbpe) and amyq (pt27-amyq). The consensus promoter and the downstream riboswitch was fused (ON10/ON19; template pt23) to bgab (pt25), htpg (pt28-htpg), pbpe (pt28-pbpe) and amyq (pt28-amyq) as well. Insertion of the wild-type promoter without (ON10/ON19, template pt12) and with the signal sequence of amyq (ON20/ON21; template pndh37) into pht01 resulted into pt30 (Fig. 2.1 A) and pt32 (Fig. 2.1 B), respectivety. Replacement of the wild-type promoter by the consensus promoter resulted in pt31 (ON10/ON19; template pt23) and pt33 (ON20/ON21; template pndh37), respectively [116]. A B Fig Genetic and restriction map of the expression vector pt30 and the expressionsecretion vector pt32. (A) pt30 with the wild-type P gcv promoter; (B) pt32 with the wildtype P gcv promoter fused to the coding region for the signal sequence. 39

48 2.7.2 Construction of expression vectors based on the lysine riboswitch To check for the function of the riboswitch, it was transcriptionally fused to the lacz gene present in the integration vector pdg1728 [52]. The region coding for the riboswitch including its promoter was generated by PCR using the primer pair ON33/ON34 and DNA of strain 1012 as template. The amplicon was cleaved with EcoRI and BamHI and ligated into pdg1728 treated with the same enzymes. The resulting plasmid pt13 was transformed into the B. subtilis strain AM01 selecting for spectinomycin-resistant cells and screening for chloramphenicol sensitivity, and strain PT42 was kept for further studies. The transcriptional fusion was also transformed into strain PT41 yielding PT43. To study the influence of the transcriptional terminator on the regulation of lysc, the region corresponding to the right arm of the inverted repeat was removed using the primer pair ON33 and ON35 resulting in pt41. ON35 contains the SD sequence of lysc to ensure translation of lacz. pt41 was transformed into strain AM01 and all following steps were as described for pt13. Strain PT44 is a derivative of AM01 and PT45 of PT41, both carrying the transcriptional fusion with the partial deletion of the terminator. To test whether the lysine system can be used as an auto-inducible expression system, plasmid pt02z was constructed allowing expression of the lacz gene. The coding sequence of lacz gene was amplified using ON38/ON39 with pmutin-ydrb as template, cleaved with XbaI and BamHI and inserted into phcmc01 treated with the same enzymes resulting in pt05z. Next, the terminator t 0 was amplified using ON40/ON41 with pbgab as template and inserted into pt05z at NheI and KpnI sites resulting in pt05-lacz (Fig. 2.2). The region coding for the riboswitch including its own promoter was generated by PCR using the primer pair ON36/ON37 and DNA of strain 1012 as template. The amplicon was fused with the lacz gene from pt05-lacz at the BclI and XbaI restriction sites resulting in pt02z. The correct plasmid was then transformed into B. subtilis 1012, and the expression level of β-galactosidase was measured in the absence and presence of different concentrations of L-lysine Construction of structurally stable plasmids It has been observed that the two plasmids pndh33 and pndh37 [117] exhibit some structural instability in E. coli [105]. When the complete DNA sequence of these two plasmids was analysed, an 117-bp sequence was detected occurring twice. This sequence in derived for the 3 end of the laci gene and occurs as a direct repeat. To increase the stability 40

49 of these two plasmids in E. coli, one copy of the 117 bp direct repeat containing the 3' end of laci was removed as follows. The DNA region containing the bla gene and the origin of replication derived from the pbr322 vector was amplified using oligonucleotides ON42 and ON43 with pmtlbs72 as template. The resulting 2010-bp amplicon was cleaved with XhoI and AflII and ligated with pndh33 or pndh37 treated with the same enzymes resulting pht01 and pht43, respectively (Fig. 2.3). Fig Genetic and restriction map of the expression vector pt05-lacz. This vector has been derived from phcmc01 containing the lacz reporter gene and the t 0 terminator of phage λ. To test whether the removal of the repeated region has any influence on the control of gene expression, the following plasmids were constructed. The coding sequences for BgaB, HtpG, Pbp4* and α-amylase AmyQ containing its own signal sequence were amplified using ON44/ON45 with px-bgab [79], ON46/ON47 and ON48/ON49 with chromosomal DNA of B. subtilis 1012 and ON50/ON51 with pkth10 [112] as template, cleaved with BamHI and AatII and inserted into pht01 treated with the same enzymes resulting in pht01-bgab, pht01-htpg, pht01-pbpe and pht43-amyq respectively. The genes cela and celb were amplified using ON52/ON53 with pct105 and ON54/ON55 with pct208 as template [30]. The amplicons were cleaved with BglII/AatII for cela and BamHI/AatII for celb and inserted into pht43 treated with BamHI and AatII resulting pht43-cela and pht43-celb, respectively. 41

50 Fig Genetic and restriction map of the expression vector pht01. Indicated are three orfs (orf-1, orf-2 and orf-3) derived from the Bacillus plasmid pbs72, the ampicillin (Amp) and chloramphenicol (Cm) resistance marker active in E. coli and B. subtilis, repectively, the replication region (rep) of pmb1, the laci gene and the P grac promoter followed by a MCS and a strong transcription terminator indicated as bar. Below the plasmid map, the DNA sequences of the MSC of pht01, of the His-tag of pht08, of the Strep-tag of pht09, the MCS preceding the Strep-tag in pht24 and the c-myc-tag in pht10 and the MCS of the expressionsecretion vector pht43. To allow detection and rapid isolation of recombinant proteins, the coding region for three different tags were inserted into pht01. We have chosen the 10-amino-acid epitope-tag c-myc [39], and the 8-amino-acid His [64] and Strep [73] purification tags. First, the coding regions for the His- and Strep-tag including the start codon were inserted into the BamHI site of pht01 downstream of the SD sequence using the two pairs of complementary oligonucleotides ON56/ON57 and ON58/ON59, respectively, resulting in the novel plasmids pht08 and pht09. The coding regions for c-myc and Strep including a stop codon allowing 42

51 their fusion to the 3 end of any gene of interest were inserted into pht01 at the AatII site using two pairs of complementary oligonucleotides each (ON60/ON61 for c-myc and ON62/ON63 for Strep) resulting in pht10 and pht24, respectively (Fig. 2.3). Next the gene yhcs was amplified using ON64/ON65, with B. subtilis 1012 chromosomal DNA as template and the core region of srta (termed as srta) of L. monocytogenes [125] using ON65/ON67 and pndh09 [106] as template; the amplicons were cleaved by appropriate enzymes and then inserted into pht08 yielding pht08-yhcs and pht08-srta. The gene ywbn and gfp+ were amplified using ON68/ON69 with B. subtilis 1012 chromosomal DNA and ON70/ON71 with pmutin-gfp+ [75] as template and inserted into pht10 and pht24 resulting pht10-ywbn, pht10-gfp and pht24-gfp. The gene gfp+ was also amplified using ON70/ON72 and then inserted into pht09 resulting pht09-gfp Construction of the promoter-probe vector pht06 The promoter-probe plasmid, pht06, basal on bgab, was constructed for the identification and screening of promoters was constructed as follows. The BgaB-encoding region was amplified by PCR using plasmid px-bgab [79] as template with primers ON77 and ON78. The amplified fragment was cloned into the pht01 [105] at KpnI/AatII site resulting pht02. The laco sequence with restriction sites for SpeI and SacII was inserted at the BamHI restriction site using the two complementary oligonucleotides ON79 and ON80 resulting in pht06 (Fig. 2.4). E. coli or B. subtilis carrying the promoter-probe plasmid pht06 will give white colonies on IPTG-X-gal LB plates Construction of plasmids to identify elements of strong promoters Construction of plasmids pht59 to pht100 (Tab. 2.2): To generate a library of promoters, core promoters and some full-length promoters were introduced into plasmid pht06 between SpeI and SacII using two complementary oligonucleotides (see on Tab. 2.3). Some upstream regions of core promoters were introduced into plasmids pht70 or pht80 between KpnI and SpeI; the number of plasmids corresponds to the number of the oligos used (Tab. 2.2 and 2.3). As examples, promoter P88 was assembled from the two complementary oligonucleotides P88F and P88R, ligated into plasmid pht06 at SpeI and SacII resulting in plasmid pht88. Transformants were first screened by blue/white colonies and then verified by colony PCR. Other synthetic promoters were introduced into this plasmid at the MCS by similar approaches. 43

52 Fig Genetic and restriction map of the promoter-probe plasmid pht Construction of plasmids allowing the analysis of 3 stabilizing elements To investigate the influence of 3 -stablizer elements on the expression of the bgab reporter gene, plasmid pht36 was constructed by removing the trpa transcriptional terminator from pht01-bgab. The coding regions for ORF-3 and ORF-1 were amplified by PCR using pmtlbs72 as template and the primer pair S36F and S36R. The AatII/ClaI-treated PCR fragment was cloned into the 7917 bp AatII/Bsp119I-treated fragment of pht01-bgab resulting in pht36 (Fig. 2.5). pht01-bgab P grac bgab ORF-3 ORF-1 pht36 trpa terminator AatII P grac bgab ORF-3 ORF-1 Terminator Fig Genetic features of the expression vector pht01-bgab and plasmid pht36 to investigate 3 stabilizer elements. 44

53 2.7.7 Construction of plasmids to study mrna stabilizing elements Construction of a plasmid with 3 stabilizing elements: Two complementary oligonucleotides were hybridized to generate a terminal stabilizing element and these terminators were introduced into plasmid pht36 at AatII resulting in plasmids pht110 to pht124 with numbers corresponding to the numbers of their primers (Tab. 2.2 and 2.3). As an example, the element S124 was generated by hybridization of the two complementary oligonucleotides S124F and S124R; this hybridized product was ligated into pht36 at AatII. The correct transformants were screened by using colony PCR, and candidate plasmids were further analysed by restriction enzymes and DNA sequencing. Construction of other plasmids with names numbered larger than pht100 (Tab. 2.2): Other promoters with stabilizing elements numbered larger than 100 (Tab. 2.3) were generated by PCR and then introduced into pht06 at the MCS region. As an example, P212 promoter with S212 was generated by two PCR steps. First, the promoter groe with laco fragment, called P groe _211* was amplified using the two primers ON76F and S211R1 and the promoter P grac in pndh33 as template. The PCR products were then purified by the gel extraction kit (Qiagen). Second, the promoter with S212 was amplified using two oligos ON76F and S212R2. The PCR fragments were digested with KpnI and BamHI and then ligated into pht06 treated with the same enzymes, resulting in pht212. The right transformants were also screened by blue/white colonies and using colony PCR, candidate plasmids were further analysed by restriction analysis and sequencing (see Fig. 2.6 for more detail) Construction of the knockout strains to study glycine expression system To delete the gcv structural genes (gcvt, gcvpa, gcvpb) from the B. subtilis chromosome (Fig. 2.7 A), about 300-bp-regions up- and downstream of the operon were amplified using ON5/ON6 and ON7/ON8 and chromosomal DNA of strain 1012 as template, and the two amplicons were used to sandwich a neomycin resistance marker as shown in Fig. 2.7, and all fragments were cloned into pbluescript 2KT (+) resulting in plasmid pt20 (Fig. 2.7 B). The recombinant plasmid was used to transform strain 1012, neomycin-resistant colonies were selected and several candidates were checked for correct replacement of the gcv operon by PCR and by the presence of the neomycin resistant marker. One positive strain, PT05, was kept for further studies. The gcv::neo marker was transferred into AM01 resulting in strain PT02 to facilitate subsequent isolation of integrants after transformation with the pdg1728 derivatives. 45

54 ON76F PCR 1 P grac S211R1 ON76F PCR 2 PgroE_211* S212R2 Stabilizing element S212 P groe _S212 P groe WT laco RBS KpnI Transcriptional start site Spacer BamHI Fig Schematic representation of the construction of the promoter PgroE_S212, P212. Positions of transcriptional start site, stabilizing element, spacer and ribosome binding site are indicated; grey box represent the laco operator. To completely delete the gcv operon (including the riboswitch and the structure genes) from B. subtilis chromosome, an about 300-bp region upstream of the gcv operon (on yqhh) (Fig. 2.7 A) was amplified using ON23/ON24 and chromosomal DNA of strain 1012 as template, and the amplicon was inserted into pt20 between the SacI and BamHI restriction sites resulting in plasmid pt37 (Fig. 2.7 C). The neomycin resistance marker was amplified using ON23/ON8 and plasmid pt37 as template, and all fragments were used to transform strain AM01. Neomycin-resistant colonies were selected and checked for correct replacement of the whole gcv operon by PCR and by the presence of the neomycin resistant marker. One positive strain, PT40, was kept for further studies. 46

55 A B C Fig. 2.7 Genetic and restriction map of the gvc operon and of the plasmids pt20 and pt37 which have been constructed for the deletion of the gcv operon. (A) The coding regions of the gcv operon on B. subtilis chromosome, gcvpb (glycine decarboxylase subunit 2), gcvpa (glycine decarboxylase subunit 1) and gcvt (aminomethyltransferase) are shown; (B) plasmid pt20 used to delete the gcv structural genes (gcvt, gcvpa, gcvpb) from the B. subtilis chromosome; (C) plasmid pt37 used to delete the whole gcv operon including gcv-riboswitch and the gcv structural genes from the B. subtilis chromosome Construction of lysc knockout strains to study lysine expression system The complete lysc gene including its leader region was deleted from the chromosome as follows. The lysc gene is flanked by the genes yslb and uvrc (Fig. 2.8 A); about 300 bp from both genes were amplified separately using the oligonucleotide pair ON29/ON30 for uvrc and ON31/ON32 for yslb and chromosomal DNA of B. subtilis strain 1012 as template. Both 47

56 amplicons were treated with the enzymes SacI and BamHI for uvrc and KpnI and XhoI for yslb, and the DNA fragments were cloned into plasmid pt20 flanking the neo gene resulting in pt39. In the last step during the construction of the lysc knockout, the flanking regions of lysc including the neo marker were amplified using ON29 and ON32 and plasmid pt39 as template. The linear DNA fragment was transformed into strain AM01, and transformants were selected on LB plates containing neomycin. Chromosomal DNA was isolated out of several neomycin-resistant transformants and analysed for replacement of the lysc gene by the neo gene by PCR. All of them turned out to correct, and one transformant (strain PT41) was kept for further studies (Fig. 2.8 B). A B. subtilis 1012 B B. subtilis PT40 uvrc neo yslb Fig Display of the chromosomal regions of the knockout strains PT41. (A) Chromosomal region of the lysc gene in B. subtilis 1012; (B) construction of strain PT41. 48

57 3 Results This work was focused on the construction of novel controllable expression systems and to identify features important for overproduction of recombinant proteins in B. subtilis. First, the L-glycine and L-lysine amino acid-responsive riboswitches that regulate gene expression are used to construct glycine-inducible and lysine-autoinducible expression vectors; and IPTGinducible expression plasmids that allow expression and purification of proteins were also constructed. Second, a protocol with a useful promoter-probe plasmid to analyze strong promoters in B. subtilis was established. Using this new technique, promoter elements and mrna stabilizing elements were studied to enhance the transcriptional level and mrna stability leading to higher protein production levels. 3.1 Exploring glycine controllable expression systems This part describes the development of a glycine-inducible expression system for B. subtilis. This is the first report that a naturally occurring riboswitch can be used for controllable overproduction of recombinant proteins using the inexpensive inducer L-glycine. First, the presence of a small transcript corresponding to the 5 UTR in the absence of L-glycine which is converted into the full-length transcript after addition of the amino acid was confirmed by Northern blot. Next the lacz reporter gene was fused with the promoter and the riboswitch and glycine-dependent induction was demonstrated. Furthermore, to obtain higher levels of recombinant proteins, the promoter strength was enhanced, and the HtpG, Pbp4* and α-amylase were used as model proteins The glycine degrading gcv operon is strongly induced after addition of L-glycine Demonstration of the transcription attenuation model of the three genes gcvt-gcvpagcvpb in-vivo The model of transcription attenuation suggested by Mandal and coworkers [93] predicts the synthesis of a short transcript in the presence of low L-glycine concentrations, while the fulllength mrna should appear after raising the concentration to 10 mm. To prove this prediction, B. subtilis 1012 cells were grown in minimal medium to the mid-logarithmic growth phase in the complete absence of L-glycine at 37 o C. Then, L-glycine was added to a final concentration of 10 mm and growth was continued. Aliquots were withdrawn immediately before adding L-glycine and at different time points after addition of this amino acid. Total RNA was prepared and subjected to Northern blotting, which was probed with two different antisense RNAs, one complementary to the riboswitch and the second to the gcvt 49

58 gene, the first gene of the tricistronic operon. When antisense RNA complementary to the riboswitch was used to probe the Northern blot, a short transcript with a length of about 200 nucleotides was predominating, while three additional very faint larger bands were present, too (Fig. 3.1A). Already 5 min after increasing the L-glycine concentration, the full-length transcript could be detected (about 4 kb) including two additional smaller bands (Fig. 3.1A). These could arise either from premature transcription termination, from internal processing, they could represent degradation products or a mixture of these possibilities. If the second probe, complementary to the gcvt gene, was used the short transcript was not detected as to be expected, while the overall pattern was comparable (Fig 3.1B). Most interestingly, the amount of the band representing the attenuation product decreased slowly over time followed by a further increase (Fig. 3.1A). One possibility to explain this surprising result could be that the attenuation product is very stable. Fig Northern blot analysis of the gcv operon. Total RNA was prepared from B. subtilis wild-type strain 1012 grown in minimal medium at 37 o C before (lane 1) and 5, 10, 30, 60 and 90 min (lanes 2-6) after addition of 10 mm L-glycine. The Northern blot was probed with antisense RNA complementary to the riboswitch (A), the gcvt gene (B), and dnak (C) which served as a loading control. 20 micrograms of RNA were applied per lane. Transcript size was determined by comparison with RNA size marker as indicated To analyse for enhanced stability of the gcv riboswitch RNA To examine this possibility experimentally, cells were grown to the mid-exponentially growth phase, and de novo synthesis of RNA was inhibited by the addition of rifampicin. As shown in Fig. 3.2B, the attenuation product was undetectable after 2 min indicating that an extended half-life does not explain the appearance of this band in Fig. 3.1A. Another possibility to explain this unexpected finding could involve processing of the full-length transcript. 50

59 5S + trna Fig Northern blot analysis for enhanced stability of the gcv riboswitch RNA. Cells of B. subtilis strain 1012 were grown in minimal medium at 37 o C to the mid-exponential phase. Then, rifampicin was added at a final concentration of 100 μg/ml to inhibit further transcription. Total RNA was prepared just before addition of rifampicin (lane 0) and 2, 4, 6, 10, 15, 20 and 30 min (lanes number 2-30) after addition of rifampicin. (A) The total RNAs were separated on an agarose gel and stained with Et.Br, and the ribosomal RNAs served as a loading control; (B) the Northern blot was probed with antisense RNA complementary to the riboswitch To analyse for processing of the full-length transcript To examine this possibility, the rifampicin experiment was repeated in a slightly different way. Cells were grown first in absence and then in presence of L-glycine, and rifampicin was added 2.5 min later. Total RNA was prepared 30 min before addition of the amino acid (t = -30), immediately after supplementation with L-glycine (t = 0) and at different time points after addition of rifampicin as indicated in Fig The Northern blot was probed with anti-riboswitch (Fig. 3.3A) and anti-gcvt RNA (Fig. 3.3B), and the X-ray film was overexposed to detect low abundance transcripts, too. While the full-length transcript and its putative degradation products could be detected already in the absence of L-glycine as described above, there was a dramatic increase upon addition of L-glycine, and both the full-length and the riboswitch RNA decayed rapidly after inhibition of transcription (Fig. 3.3A). When the Northern blot was probed with anti-gcvt, the result was different. Here, the full-length RNA was more stable (Fig. 3.3B). While no full-length RNA could be detected 20 min after addition of rifampicin when the blot was probed with antiriboswitch RNA, this transcript was still present, though in reduced amounts, using anti-gcvt RNA (compare Fig. 3.3A and 3.3B). The only interpretation of this result is that indeed processing of the full-length transcript occurs to remove the riboswitch RNA. It follows that only the high molecular weight transcript in the Fig. 3.3A represents full-length RNA, while the corresponding transcript in the Fig. 3.3B corresponds to a processing product. The 51

60 difference in molecular weight between these two RNA species cannot be resolved under the experimental conditions used here. In summary, the continued synthesis of the riboswitch RNA neither results from an enhanced half-life nor from processing of the full-length transcript. Fig The gcv encoded attenuation product is rather unstable and its continued synthesis during the induction phase does not result from processing. A B Cells of B. subtilis strain 1012 were grown in minimal medium at 37 o C to the midexponential phase, cells were induced with glycine and then rifampicin was added at a final concentration of 100 μg/ml to inhibit further transcription. Total RNA was prepared 30 min before addition of L-glycine (t = -30), immediately after addition of the inducer (t = 0) and rifampicin (t = 2.5) at the time points indicated. The Northern blot was probed with antisense complementary to the riboswitch (antiriboswitch) (A) and with anti-gcvt RNA (B) Detection of the DNA sequence coding for the transcriptional terminator results in the disappearance of the riboswitch RNA Base on published data [93], it can be expected that removal of the transcriptional terminator located at the 3 end of the riboswitch will result in constitutive expression of gcv operon independent of the absence or presence of L-glycine in the growth medium. But will the riboswitch RNA be synthesized in such a deletion mutant? To answer this question, the coding region of the terminator was deleted as described in the Methods section 2.7.1, and the deletion derivative was fused to the lacz reporter gene and integrated at the amye locus in strain PT40 with a complete deletion of the gcv operon (strain PT49). As a control, the wildtype promoter-riboswitch region was fused to lacz and integrated ectopically in the same strain (PT47). 52

61 Next, both strains were incubated in minimal medium. When the cultured cells reached the mid-log growth phase, L-glycine was added as described. Aliquots were withdrawn just before adding of the L-glycine (t = 0) and at different points after induction and probed with anti-riboswitch RNA. When the operon fusion carrying the wild-type riboswitch was analysed, only the leader transcript was present when the cells grew in the absence of L-glycine. Addition of L-glycine resulted in the appearance of additional bands as described above (Fig. 3.4A). Next, expression was analysed with a truncated the transcriptional terminator. As to be expected expression was constitutive and the riboswitch RNA was not produced (Fig. 3.4B). These results clearly indicate that the terminator plays an important role in transcription termination and release of the riboswitch RNA, and the gcv-riboswitch is also synthesised if fused to a foreign gene. This observation opens the possibility to use the gcvriboswitch for controlled expression of recombinant gene. Fig In the absence of the transcriptional terminator the riboswitch RNA is absent. (A) Strain PT47 carrying the transcription terminator; (B) strain PT49 devoid of part of the terminator. The two strains were first grown in minimal medium at 37 o C to the midexponential phase (t = 0), cells were induced with 10 mm L-glycine and then total RNA was prepared at the time points indicated and probed with anti-riboswitch RNA Expression of the reporter gene lacz fused to glycine riboswitch To determine the optimal L-glycine concentration resulting in highest induction factor Based on these results, the promoter activity was measured in the absence and presence of L-glycine. The lacz gene was fused to the riboswitch including the σ A -dependent promoter, and the transcriptional fusion was integrated into the B. subtilis chromosome at the amye locus using strain PT02 as recipient which carries a deletion of the gcv operon resulting in PT46 as described in the Methods section

62 First, the different L-glycine concentrations (from 0.1 to 100 mm) were tested to determine the optimal concentration resulting in the highest induction factor. Cells were grown in minimal medium to the mid-exponential growth phase. Then, the culture was split into six subcultures, where one was incubated without further treatment, while the others were induced with different L-glycine concentrations as indicated in Fig Aliquots were withdrawn from all cultures for determination of the enzymatic activity. 50 mm L-glycine resulted in the highest β-galactosidase activities, which increased dramatically within the first hour after addition of L-glycine to about 48 units followed by a further increase to approximately 65 units 6 h after induction. Since the difference in the enzymatic activity at later times between 10 and 50 mm was only minor (Fig. 3.5), the 10 mm of L-glycine was used in all subsequent experiments. Fig Full induction of the glycine-dependent riboswitch RNA occurs after addition of a L-glycine concentration of mm. Cells of strain PT46 were grown in minimal medium in the absence of L-glycine at 37 o C to the mid-exponential growth phase. Then, the culture was split into six subcultures, where one was further grown in the absence and the others in the presence of L-glycine at the concentrations indicated. Aliquots were removed at the time points indicated for determination of the β-galactosidase activities. 54

63 The effect of culture medium on β-galactosidase activity To follow induction more closely and to compare the increase in β-galactosidase activity in minimal and LB medium, the experiment was repeated. The β-galactosidase activity started to increase slightly 10 min after addition of L-glycine and reached the maximum level after about 2 h in LB medium (Fig. 3.6A). Since the enzymatic activity of the culture grown in the absence of L-glycine also started to increase after about 60 min, an induction factor of 2.3 can be calculated after 240 min (Fig. 3.6A). When the same experiment was carried out in minimal medium the induction levels were higher but reached about the same values after 2 to 4 h (Fig. 3.6B). In contrast to the culture grown in LB medium, the β-galactosidase activity did not increase over time resulting in an induction factor of 10. Fig The riboswitch of the gcv operon confers glycine-inducibility to the lacz reporter gene. The transcriptional fusion present in plasmid pt12 was integrated at the amye locus in strain PT02 which carries a deletion of the gcv operon. Cells were grown in LB (A) or in minimal medium (B) to the mid-exponential growth phase. Aliquots were withdrawn immediately before (t = 0) and at the time points indicated after addition of 10 mm L-glycine. The enzymatic activities are expressed in relative β-galactosidase units. The reporter enzyme activities were measured from cultures without added L-glycine (open bars) and at the time point indicated after addition of 10 mm L-glycine (closed bars). 55

64 These results prove that glycine-inducible expression of lacz can be obtained Does the growth temperature influence the expression level? Next, I asked whether the growth temperature will influence the basal level of expression through the stability of the attenuation terminator. The stability of RNA secondary structure is temperature dependent. This is used to regulate initiation of translation by several genes such as the rpoh gene of E. coli, the ROSE element of Bradyrhizobium japonicum and the prfa gene of L. monocytogenes [26, 26, 72, 72, 102, 102]. To find out whether the growth temperature will influence the basal expression level or the expression level after increasing the L-glycine concentration or both, cells were grown at three different temperatures, and the β-galactosidase activity was measured. Since the cells were grown up to 45 o C, lacz could not be used as a reporter gene since its activity is highly unstable at high temperatures [177]. Instead, we fused the bgab gene to the riboswitch coding for the heat-stable β-galactosidase (pt24) [62], and the fusion is expressed from a plasmid assumed to be present in about four copies [156]. Cells were grown at three different temperatures (30, 37 and 45 o C) in LB medium till the mid-logarithmic growth phase was reached (t = 0). Then, the cultures were divided while one subculture was further grown without addition of L-glycine while 10 mm L-glycine was added to the second. The basal levels measured after 2 h in all three uninduced control cultures grown at three different temperatures did not vary considerably, and the induced levels remained comparable in the 30 and 37 o C cultures, too (Fig. 3.7A and B), but was reduced in the 45 o C culture (Fig. 3.7C). These results revealed that growth from 30 to 45 o C did not influence the basal level of expression, while the high temperature reduced the induced level Analysis of mutant gcv promoters A comparison of the DNA sequence of the σ A -dependent promoter directing expression of the gcv operon with that of the consensus sequence [57, 58] revealed that the -35 region deviates in four and the -10 region in only one position from the consensus sequence (Fig. 3.8). In addition, there is an extended -10 region present. Such variants of the -10 element have been reported to increase the strength of the promoter, and they are often accompanied by a weak -35 element as shown in the present case [57]. 56

65 Fig Influence of the growth temperature on the basal and induced level of the reporter enzyme. Cells were grown in LB medium at 30 o C (A), 37 o C (B) and 45 o C (C). Aliquots were withdrawn for β-galactosidase determinations before (0 h) and at different times after addition of 10 mm L-glycine (1 to 4 h) from the uninduced (open bars) and induced (closed bars) cultures. Fig DNA sequences of the wild-type gcv promoter, the consensus sequence of the σ A housekeeping promoter and of three promoter mutants. TG represents the extended -10 region. 57

66 The question was asked whether altering the two elements of the gcv promoter will lead to a stronger expression level. To this end, both the -35 and the -10 regions were mutated separately and both elements together to the consensus sequence as shown in Fig All three strains were grown in LB and in minimal medium (only the data for minimal medium are shown) as described in 2.3.3, and L-glycine was added when the cells had reached the mid-log growth phase. Aliquots were taken for determination of β-galactosidase activities as shown in Fig When the promoter with the -10 consensus region was analyzed, the basal levels were comparable with those measured with the wild-type promoter (compare Figs. 3.6 and 3.9A). Addition of 10 mm L-glycine led to reduced expression suggesting that the gcv promoter with a -10 consensus sequence is weaker relative to the wild-type promoter. At least, the mutation to the consensus sequence did not improve the promoter strength. When the promoter with the -35 consensus element was analyzed both the basal and the induced levels were significantly increased, and an induction factor of about 10 was measured (Fig. 3.9B). In summary, the alterations in the -35 region improved the promoter strength, but at the expense of the basal level. When both the -35 and the -10 regions were changed to the consensus sequence, the basal level was increased more than 10-fold as compared to the wildtype promoter (compare Figs. 3.6 and 3.9C), and the induced level was slightly decreased as compared to the promoter with the -35 consensus sequence, leading to an induction factor of about 3.5. In conclusion, while the 10 consensus element did not led to a promoter improvement, the 35 consensus increased the promoter strength about 5-fold Fusion of the groe promoter to the riboswitch Can the gcv promoter be replaced by a foreign promoter without affecting the induction behaviour? The strong groe promoter [117] was fused to the riboswitch and to lacz and the operon fusion was integrated at the amye locus resulting in PT17 (see 2.7.1). As already observed for the consensus gcv promoter, the basal level was rather high, and the induced level reached values about 10-fold higher after 3 h of induction (Fig. 3.10A). In conclusion, the promoter can be exchanged without impairing the induction behaviour. In another experiment, the gcv promoter-riboswitch region was moved to the plasmid pht01 fusing it to the bgab reporter gene (pt24). Here, the results from Figs. 3.7B and 3.10B were compared; in both cases, cells were grown at 37 o C, bgab was used as reporter gene and the fusion is expressed from a plasmid assumed to be present in about four copies [156]. In the first case, cells were grown in LB medium and the second case is minimal medium. The basal level was significantly lower when the reporter gene was expressed in LB medium. 58

67 Fig Analyses of mutant gcv promoters. The β-galactosidase activities expressed from promoters with the -10 consensus element (A), the -35 consensus element (B) and both consensus elements (C) were analyzed with cells grown in minimal medium. See legend to Fig. 3.6 for further explanations. Fig Expression of reporter genes fused to the groe promoter integrated in the chromosome and to the gcv promoter located on a plasmid. (A) The groe promoter was fused to the riboswitch and to lacz and the operon fusion was integrated at the amye locus, see legend to Fig. 3.6 for further explanations; (B) the wild-type gcv promoter was fused to the bgab gene present on a plasmid. Cells were grown in minimal medium; see legend to Fig. 3.7B for further explanations. A B 59

68 3.1.6 Synthesis of recombinant proteins from plasmid-based expression systems In the last series of experiments, the amount of several recombinant proteins were analyzed which can be produced using the glycine riboswitch with the wild-type (pt27) and the consensus promoter (pt28) using the vector plasmids pht01 and pht43. As model proteins, we have chosen two intra- and one extracellular protein (see in 2.7.1) Synthesis of intracellular recombinant proteins The intracellular protein is encoded by htpg heat shock gene [143] and pbpe coding for a penicillin binding protein [118]. Synthesis of both proteins was analyzed in LB and in minimal medium, before and 2 and 4 h after glycine-induction by separation of the whole proteins on an SDS-PAGE followed by Coomassie staining (Figs and 3.12). Fig Identification of the htpg gene product. B. subtilis strain PT05 carrying either pt27-htpg or pt28-htpg were grown in either minimal (A) or LB medium (B) at 37 o C to midlog, divided into two subcultures, where one was further grown untreated and the second treated with 10 mm L-glycine. Aliquots were taken before addition of L-glycine (0 h), two hours later from the untreated (2 h (-)) and 2 and 4 h later from the induced cultures (2 h and 4 h). Cells were lysed by lysozyme and 100 μg of protein were loaded per lane on an 10% SDS-PAGE. After gel electrophoresis, the proteins were stained with Coomassie blue. Immunoblot analysis of HtpG present in the cells grown in minimal (C) and LB (D) medium. 60

69 While no HtpG-specific and Pbp4*-specific band could be detected when the gene was fused to the wild-type promoter in both minimal and LB medium after addition of L-glycine (Figs. 3.11A, B and 3.12A, B), a band with the molecular weight of HtpG and Pbp4* became visible with the consensus promoter in both media (Figs. 3.11A, B and 3.12A, B). To use a more sensitive method allowing detection of even small amounts of protein, immunoblots were carried out. Here, HtpG and Pbp4* were shown to be present in all samples even before induction, but its amount increased significantly after addition of L-glycine (lanes 2 h and 4 h in Figs. 3.11C, D and 3.12C, D). It has to be mentioned that there is a low level expression of htpg and pbpe from its chromosomal location. The highest amount of HtpG and Pbp4* could be detected when the gene was transcribed from the consensus promoter and the cells were cultivated in minimal medium. These data are in agreement with measurement of the β- galactosidase activities. A B Fig Identification of the pbpe gene product. B. subtilis strain PT05 carrying either pt27-pbpe or pt28-pbpe were grown and treated as described in the legend to Fig Coomassie blue staining of Pbp4* present in cells grown in minimal (A) and LB medium (B). Immunoblot analysis of Pbp4* present in the cells grown in minimal (C) and LB (D) medium Synthesis of extracellular recombinant proteins Synthesis of extracellular proteins coded for by the amyq gene [112] was analyzed in minimal and in LB medium, before and 2 and 4 h after glycine-induction. Production of 61

70 α-amylase was analyzed first by checking for halo formation on minimal and LB medium containing starch. As to be expected small colonies were formed on minimal and larger ones on LB medium (Fig. 3.13). The size of the halo was influenced by the promoter used and by the absence or presence of L-glycine. It turned out that the largest amount of α-amylase was secreted when the amyq gene was fused to the consensus promoter in the presence of 10 mm L-glycine (Fig. 3.13). Next, this result was verified by applying equal amounts of protein collected from the supernatants of the two strains grown in the two media to SDS-PAGE analysis. As can be seen from Fig. 3.14, α-amylase was produced in both media and from both promoters, but the largest amount of enzyme was secreted in LB medium using the consensus promoter (Fig. 3.14B) which could be confirmed by Western blots (Fig. 3.14D). Fig Production of α-amylase by single colonies as verified by halo formation on plates containing starch. Single colonies were analyzed on either LB or minimal medium in the absence or presence of 10 mm L-glycine. These results clearly demonstrate that the glycine riboswitch can be used for regulatable production of both intra- and extracellular proteins. Four plasmid-based expression vectors have been constructed where two allow intracellular production of recombinant proteins (pt30 and pt31), while the other two direct the proteins into the culture medium (pt32 and pt33) (see in 2.7.1). Both vectors pt30 and pt32 use the wild-type P gcv promoter while the other two pt31 and pt33 use the consensus promoter allowing their induction by addition of L-glycine (Fig. 2.1). 62

71 A B C Fig Identification of α-amylase. Cells carrying the plasmids pt27-amyq or pt28-amyq were grown in either minimal (A) or LB medium (B) in the absence (0 h, 2 h (-)) or after addition of 10 mm L-glycine (2 and 4 h). Cells were pelleted by centrifugation and equal amounts of supernatant were analyzed by a 10% SDS-PAGE stained with Coomassie blue. In addition, the α-amylase produced was visualized by immunoblotting in minimal medium (C) and in LB medium (D). D 3.2 Exploring lysine controllable expression systems This part describes the development of a lysine auto-inducible expression system for B. subtilis. First, a detailed transcriptional analysis of the lysine-responsive and riboswitchregulated lysc operon of B. subtilis, the presence of a small transcript corresponding to the 5 UTR in a high concentration of L-lysine which is converted into the full-length transcript after remove or in the presence of a low concentration of the amino acid was confirmed and carried out by Northern blot. Next, the lacz reporter gene was fused with the promoter and the riboswitch and lysine-dependent induction was demonstrated. Furthermore, it was shown that using system can be used as the auto-inducible expression system based on different L-lysine concentrations. 63

72 3.2.1 Transcriptional analysis of the lysc gene As reported above, expression of the lysc gene is regulated by a lysine-responsive riboswitch which directly senses the amount of L-lysine present within the cell [51, 153]. To test for the switch in transcription following removal of L-lysine, a Northern blots was carried out using two different antisense RNAs, one which detects the riboswitch and the full-length transcript and the other the full-length transcript only. B. subtilis strain 1012 was grown in minimal medium supplemented with 300 μg/ml of L-lysine. When cells reached the mid-exponential growth phase, they were sedimented by centrifugation, washed twice in minimal medium without L-lysine and resuspended in the original volume without L-lysine. Aliquots were withdrawn immediately before centrifugation (t = 0) and up to 180 min later. When the riboswitch probe was used, only an about 0.27 kb transcript was present when the L-lysine concentration was high (Fig. 3.15A). Upon removal of the L-lysine, the full-length 1.6-kb transcript became detectable already after 5 min with no further increase up to 180 min (Fig. 3.15A). When the same RNA preparations were challenged with anti-lysc RNA, only the fulllength transcript became apparent after removal of the L-lysine (Fig. 3.15B). In conclusion, removal of L-lysine from the growth medium changes the transcription profile from the transcription attenuation product to the full-length product within 5 min. Surprisingly, the 0.27-kb riboswitch persists for up to at least 180 min as the dominant transcript. Two different possibilities were envisaged to explain for the continued presence of the riboswitch RNA, stability or processing Analysis for enhanced stability of the riboswitch RNA To analyse for enhanced stability of the riboswitch RNA, cells of strain 1012 were grown in minimal medium to the mid-exponential growth phase and rifampicin was added to inhibit further transcription. Aliquots were removed just before addition of the antibiotic and up to 10 min later. As can be seen from Fig. 3.16B, the 0.27-kb transcript decayed rapidly with a half-life of 2-4 min. Therefore, increased stability of the riboswitch RNA under these growth conditions can be excluded. 64

73 Fig Northern blot analysis of the lysc gene of B. subtilis. Cells of strain 1012 were grown in minimal medium in the presence of 300 μg/ml L-lysine till the mid-exponential growth phase. Then, the cells were sedimented by centrifugation, washed twice in minimal medium without L-lysine and finally resuspended in the same medium. Aliquots were removed for total RNA preparation just before the centrifugation step (t = 0) and at the time points indicated. The Northern blot was probed with (A) anti-riboswitch RNA; (B) anti-lysc RNA. (C) an Et.Br stained gel is shown that comparable amounts of RNA were applied per lane. Fig The riboswitch RNA is unstable. Cells of strain 1012 were incubated in minimal medium in the presence of 300 μg/ml L-lysine. Then, rifampicin was added at a final concentration of 100 μg/ml. Aliquots were removed just before addition of the antibiotic (t = 0) and up to 10 min later. (A) Et.Br stained gel; (B) Northern blot probed with anti-riboswitch RNA Analysis for processing of the full-length transcript Next, the possibility was investigated whether the riboswitch RNA may be derived from the full-length 1.6-kb transcript by processing. In this experiment, cells were grown again in minimal medium supplemented with L-lysine followed by its removal as described. Rifampicin was added immediately after resuspension of the cells in L-lysine-free medium, and aliquots were taken 30 min before removal of the L-lysine and up to 20 min in the absence of added L-lysine. When the RNA preparations were probed with anti-riboswitch RNA, a dramatic increase in the amount of the leader transcript can be observed which 65

74 rapidly decayed over time as described above (Fig. 3.17A). When the same RNAs were probed with anti-lysc RNA, a tiny band was present before removal of the L-lysine, followed by a significant increase at t = 0 and a rapid decrease in the presence of rifampicin (Fig. 3.17B). 5 min after addition of the antibiotic, no 1.6-kb transcript is visible. This result clearly argues against processing as a possibility for the presistance of the riboswitch RNA in the absence of L-lysine. There is only one possibility and that is continued synthesis. Fig The riboswitch RNA is not produced by processing. Strain 1012 was grown as described in minimal medium with added L-lysine and cells were sedimented, washed as described and resuspended in L-lysine-free minimal medium. Then, rifampicin was added at a final concentration of 100 μg/ml at t = 0. Aliquots were removed 30 min before transfer into the L-lysine-free medium (t = -30) and at the time points indicated and hybridized to the riboswitch riboprobe (A) and lysc riboprobe (B) Deletion of the DNA sequence coding for the transcriptional terminator results in the disappearance of the riboswitch RNA It has been published that point mutations within the leader region of the lysc gene results in constitutive expression [24, 174]. Based on these results it can be expected that removal of the transcriptional terminator will also result in constitutive expression of lysc independent of the presence or absence of L-lysine in the growth medium. But what happens to the riboswitch RNA? Will it be synthesized in such a deletion mutant? To answer this question, the coding region of the terminator was deleted as described in the Methods section, and the deletion derivative was fused to the lacz reporter gene and integrated at the amye locus in strain PT41 66

75 with a complete deletion of the lysc region (strain PT45). As a control, the wild-type promoter-riboswitch region was fused to lacz and integrated ectopically in the same strain (PT43). Next, both strains were incubated in minimal medium supplemented with L-lysine. When the culture reached the mid-log growth phase, the L-lysine was removed as described. Aliquots were withdrawn just before removal of the L-lysine (t = 0) and at different time points after removal and probed with anti-riboswitch RNA. When the operon fusion carrying the wildtype riboswitch was analysed, only the leader transcript was present when the cells grew in the presence of L-lysine. Removal of L-lysine resulted in the appearance of additional bands, where the smaller ones most probably represent degradation products (Fig. 3.18A). In contrast to strain 1012 where the maximum amount of the full-length transcript became already apparent after 5 min, it took about 60 min with the lacz transcript. In addition, the riboswitch RNA was present all the times as described before. In conclusion, this result demonstrates that the riboswitch is also synthesised if fused to a foreign gene and that such a fusion can change the induction behaviour. This observation opens the possibility to use the lysine riboswitch for controlled expression of recombinant genes (see Discussion). Next, expression was analysed with a truncated transcriptional terminator. As to be expected expression was constitutive and the riboswitch RNA was not produced (Fig. 3.18B). This result clearly indicates that the terminator plays an important role in transcription termination and the release of the riboswitch RNA Fig In the absence of the transcriptional terminator the riboswitch RNA is absent. (A) Strain PT43 carrying the transcription terminator and (B) strain PT45 devoid of part of the terminator were first grown in minimal medium supplemented with L-lysine followed by centrifugation, washing and resuspension in L-lysine-free minimal medium at t = 0. Total RNA was prepared at the time points indicated and probed with antiriboswitch RNA. 67

76 3.2.3 Removal of L-lysine from the growth medium leads to an about ten-fold induction of lacz The results obtained with the Northern blot shown in Fig. 3.18A prompted me to quantify the induction of the lacz reporter gene after removal of the L-lysine. Cells of strain PT42 were grown in minimal medium supplemented with 300 μg/ml L-lysine to mid-log, washed and divided into two subcultures. While one was further grown in the presence of L-lysine, the second was incubated in its absence. Aliquots were removed for determination of β-galactosidase activities just before centrifugation (t = 0) and 2, 3 and 4 h later. While about 4 units of β-galactosidase activity were measured before centrifugation which increased to about 30 units upon further growth in the presence of L-lysine, 130 units were found after 1 h of growth in the absence of the amino acid which increased to approximately 270 units upon further growth (Fig. 3.19A). In summary, removal of L-lysine from the growth medium resulted in an about 10-fold induction of the lacz gene. A B Fig Analysis of the promoterriboswitch region fused to lacz. Strain PT42 with the wild-type lysc region (A) and strain PT43 carrying a deletion of lysc (B), and both containing the promoter-riboswitch region of lysc fused to lacz were grown in minimal medium supplemented with L-lysine, centrifuged, washed and split into two cultures as described in the legend to Fig Aliquots were removed for measurement of β-galactosidase activity at the time points indicated. 68

77 Next, I checked whether the induction behaviour will be influenced by the presence of the wild-type lysc gene. The experiment was repeated with strain PT43 carrying a complete deletion of the lysc operon including its promoter and riboswitch region. As can be seen from Fig. 3.19B, the basal level was increased from 30 to 40 units in the presence of L-lysine while removal of the amino acid resulted in 770 to 970 units. In conclusion, deletion of the wildtype lysc copy from the chromosome increased the induction factor to 25. In another experiment, to answer the question what extend inactivation of the transcription terminator will influence the expression level of the lacz reporter gene, strains PT44 and PT45 with and without the wild-type chromosomal copy of lysc and the transcriptional fusion between the promoter region and the riboswitch with the truncated terminator sequence and lacz were analysed. While close to 600 units were measured in strain PT44 in the presence of L-lysine, there was a slight increase after its removal from 720 about 1000 units in both the presence and absence of L-lysine (Fig. 3.20A). A slight increase was already seen in the wildtype background (Fig. 3.19A) and seems to be the result of growth to a higher cell density. But most importantly, there is no difference in the β-galactosidase activities between the two cultures confirming the constitutive expression of lacz in the absence of a functional transcription terminator. When the operon fusion was analysed in strain PT45, no significant difference was found as compared to the isogenic wild-type strain (Fig. 3.19B). From this result I infer that the presence of a wild-type riboswitch does not influence the expression level at a second copy with a truncated transcriptional terminator The auto-inducible expression systems From the result shown in Fig. 3.19A, removal of L-lysine from the growth medium resulted in an about 10-fold induction of the lacz gene. Next, the whole of cassette was transferred onto the promoter- probe plasmid pt05-lacz as described under Materials and methods resulting in pt02z and checked for the influence of the induction factor. B. subtilis strain 1012 carrying pt02z was grown in minimal medium with L-lysine as described, and the culture was divided into two subcultures when cells reached mid-exponential growth. Where one was further incubated in the presence of L-lysine, the other was grown in the absence of L-lysine, the β-galactosidase activities in the supplemented medium increased from 43 units at t = 0 to 56 units at t = 270 min (Fig. 3.21A). When the L-lysine has been removed, the enzymatic activity increase within the first 15 min to 270 units and reached its maximum value after 90 min with 520 units (Fig. 3.21A) resulting in an induction factor of about

78 A B Fig Analysis of the promoterriboswitch region fused to lacz in the absence of the transcriptional terminator. Strain PT44 (A) and strain PT45 (B) were grown and treated as described in the legend to Fig Next, different L-lysine concentrations (from 2.5 to 20 µg/ml) were tested to determine the optimal concentration resulting in the highest induction factor and for the purpose of creating an auto-inducible system. Cells were grown in minimal medium supplemented with 300 μg/ml L-lysine to mid-log, washed and divided into seven subcultures. Where one was further grown in the absence of L-lysine, the others were incubated in the presence of different L-lysine concentrations as indicated in Fig. 3.21B (2.5, 5, 7.5, 10, 15, and 20 µg/ml). Aliquots were removed for determination of β-galactosidase activities after centrifugation 0.5, 1, 2, 4, 6, 8, 10 and 24 h. 5 μg/ml L-lysine resulted in the high induction factor and could be used for auto-induction (Fig. 3.21B). In conclusion, a novel auto-inducible expression system for B. subtilis and related species allowing the regulated expression of recombinant genes has constructed. This system is based on the low concentration of L-lysine in the mid-exponential growth phase leading to the concomitant induction of the recombinant gene. 70

79 Fig β-galactosidase activity of lacz reporter gene located on a vector plasmid. Strain 1012 contains the vector plasmid pt02z with the wild-type lysc region (P lysc - riboswitch) fused to lacz was grown in minimal medium supplemented with L-lysine, centrifuged, washed and split into different cultures. (A) To analyse the promoter-riboswitch region fused to lacz on the vector plasmid, cells were washed and split into two cultures as described in the legend to Fig. 3.15; (B) full induction of the lysine-dependent riboswitch RNA occurs after addition of a L-lysine concentration of µg/ml. Cells were washed and divided into seven subcultures with different L-lysine concentrations (0, 2.5, 5, 7.5, 10, 15, and 20 µg/ml). Aliquots were removed for measurement of β-galactosidase activity at the time points indicated. 71

80 3.3 Construction of plasmids allowing detection and single-step purification of recombinant proteins expressed in B. subtilis Six novel plasmid-based IPTG-inducible expression vectors were constructed for B. subtilis. While one vector allows intracellular production of recombinant proteins, the second provides secretion. The third vector allows addition of the c-myc epitope tag, and the remaining three vectors provide the purification tags His and Strep. The versatility of all six vectors was proven by insertion of appropriate reporter genes and demonstration of regulated overexpression. Recombinant proteins with a His- or Strep-tag could be purified to near homogeneity in a single step Removal of a 117-bp direct repeat results in structural stability of expression vectors in E. coli During the work with the two novel expression and secretion vectors pndh33 and pndh37 [117], respectively, in E. coli, these vectors were observed to suffer from structural instability. After transformation of these two cloning vectors, small and large colonies appeared on the plates. Analysis of plasmids from small colonies revealed the expected size, while those from large colonies carried smaller plasmids allowing faster growth. To find out the reason for this structural instability, we investigated the DNA sequence of both vector plasmids for the presence of a direct repeat. Direct repeats have been reported to frequently result in deletions removing the DNA located between the direct repeats and one copy of the repeat [4]. Indeed, inspection of the DNA revealed the presence of a 117 bp segment occurring twice in the DNA and carrying the 3' end of the laci gene (data not shown), where the second copy is located between the SphI and DrdI sites in Fig. 1 of [117]. The second copy was removed as described under Materials and methods yielding the derivative plasmids pht01 and pht43 (Fig. 2.3). Both plasmids resulted in colonies of equal size after transformation into E. coli. To compare the structural stabilities of the former plasmid pndh33 with the new plasmid pht01 in E. coli, the derivatives containing the bgab reporter gene coding for a heatstable β-galactosidase was used [61]. The experiment started from one blue colony formed on LB plates containing X-Gal and 1 mm of IPTG, which were grown overnight in selective LB medium. The plasmid DNA from these overnights and from serial dilutions and new overnights were analyzed up to about 80 generations. While pht01-bgab remained stable, a smaller plasmid derivative of pndh33-bgab became apparent already after about 40 generations and predominated 72

81 thereafter (Fig. 3.22). The size of the smaller plasmid was determined to be 3.4 kb which is in perfect agreement with recombination between the two direct repeats where the smaller recombination product carries the bla gene and the origin of replication, while the bgab gene has been deleted. When cells were analyzed for their phenotype on X-Gal plates after about 80 generations, more than 99% of them exhibited a blue phenotype when carrying pht01- bgab, while only 0.2% of those inheriting pndh33-bgab showed blue colonies already after 40 generations. It was concluded that indeed the 117 bp direct repeat is responsible for the structural instability of the plasmids pndh33 and pndh37. Both, pht01 and pht43 will substitute pndh33 and pndh37 as expression and expression-secretion vectors, respectively. Fig Analysis of the structural stability of two different plasmids. Single colonies of E. coli strain XL1 Blue carrying pht01-bgab and pndh33-bgab, respectively, were grown overnight in LB medium at 37 C in the presence of ampicillin. Plasmids were prepared from these overnight cultures, cut with AflII and BglI and analyzed by agarose gel electrophoresis (lane 1). About 1,000 cells of the overnight were used to prepare a new overnight, etc. Lane 1 through 4 represent the plasmids obtained after about 20, 40, 60, and 80 generations Expression levels of the new expression vectors To compare the expression and secretion levels of the new with the previous vector plasmids, first, two reporter genes htpg and pbpe were fused to the strong Pgrac promoter in pht01 resulting in pht01-htpg and pht01-pbpe, where htpg codes for a heat shock protein of so far unknown function [61], and pbpe for the penicillin-binding protein 4* [118]. B. subtilis strain 1012 carrying these recombinant plasmids and the already described pht01-bgab were grown in LB medium at 37 C to mid-log and IPTG was added at 1 mm. Aliquots were withdrawn just before adding IPTG (Fig. 3.23A, lanes 1, 4 and 7), 2 (lane 2, 5 and 8) and 4 h (lanes 3, 6 and 9) and analyzed for the production of the recombinant proteins by SDS-PAGE and Coomassie blue staining. In all three cases, a prominent protein band can be visualized corresponding to the expected size of the recombinant protein (Fig. 3.23A). It also becomes 73

82 clear that 4 h of induction did not lead to a further increase in the amount of recombinant proteins as compared to 2 h of induction. To further verify overexpression of the recombinant proteins, after addition of the inducer, 0.15 OD578 were applied per lane except for lanes C4 to C6 where only OD578 were added. Western blots were carried out using antibodies raised against Pbp4* and HtpG; antibodies against BgaB are not available. When the Western blot was probed with αpbp4*, a very weak band could be detected in the uninduced culture (Fig. 3.23B, lane 7) which dramatically increased after addition of IPTG to the growing cells (lanes 8 and 9). A similar result was obtained when the Western blot was probed with αhtpg (compare Fig. 3.23C, lane 4 with lanes 5 and 6 where only OD578 were added). These data clearly demonstrate that the new expression vector pht01 results in overexpression of recombinant proteins comparable to the results obtained with pndh33 [117]. Fig Overexpression of three different proteins fused to the Pgrac promoter. B. subtilis 1012 cells carrying the recombinant plasmids pht01-bgab, -htpg and -pbpe were grown in LB medium at 37 C to mid-log. Then, IPTG was added at 1 mm and cells were incubated for another 4 h. Aliquots were withdrawn immediately before adding ITPG (lanes 1, 4 and 7) and 2 (lanes 2, 5 and 8) and 4 h (lanes 3, 6 and 9) after addition of the inducer OD578 were applied per lane except for lanes C4 to C6 where only OD578 were added. (A) Coomassie blue staining of the 10% SDS-PAGE; (B) and (C) Western blots probed with αpbp4* and αhtpg, respectively. 74

83 Next, the new expression-secretion vector pht43 was compared with pndh37 using three different reporter genes, cela, celb and amyq, where the first two code for cellulases and the latter for an α-amylase [30, 112]. As described before, B. subtilis cells were grown in LB medium at 37 o C to the mid-log phase (t = 0) followed by induction with 1 mm of IPTG, and the supernatants were analyzed for the presence of the secreted recombinant protein. While none of the three proteins was present in detectable amounts before addition of IPTG (Fig. 3.24, lane 0), additional protein bands started to appear 1 h after induction (lane 1) and increased up to about 5 h (lane 5). While the amounts of CelA and CelB appeared to be comparable with both vectors, significantly higher amounts of AmyQ were synthesized with the new vector pht43 (compare lanes 5 in Fig. 3.24C). In conclusion, both vectors produce at least comparable amounts of exoenzymes. Why the amount of AmyQ expressed from pht43 is higher than the amount synthesized with pndh37 is unknown. Fig Secretion of different exoenzymes. B. subtilis 1012 cells carrying either pht43-celb, -cela or -amyq were grown in LB medium as described before. Supernatant was removed just before adding 1 mm IPTG (lane 0), and 1 to 6 h later (lanes 1 to 6). 300 μl of supernatant was applied per lane of a 10% SDS- PAGE which was stained with Coomassie blue. The data for the plasmids pndh37-amyq, pndh37- celb and pndh37-cela were taken from [117]. 75

84 Incorporation of the epitope tag c-myc Preparation of monoclonal antibodies against proteins is a time-consuming task taking several months. To reduce the time involved, epitope tags have been developed which can be added either to the N- or to the C-terminus of any protein. We constructed a derivative of the vector pht01 where the coding region for the c-myc tag [39] was inserted in such a way that it can be added to the C-terminus of any recombinant protein of interest resulting in pht10. Next, the coding regions of two different genes were fused in-frame to c-myc. We have chosen the genes ywbn and gfp coding for a putative sortase [111] and for the GFP [141]. Synthesis of both proteins was induced by addition of IPTG as described before using the recombinant plasmids pht10-ywbn and pht10-gfp. While both proteins were not detectable in the absence of IPTG (Fig. 3.25A, lanes 1 and 4), they became detectable 2 h after induction of the cells with IPTG (lanes 2 and 5) and did not increase in their amount when analyzed after 4 h (lane 3 and 6) as described above for different proteins. Fig Detection of two different proteins carrying the c-myc epitope tag by immunoblotting. B. subtilis 1012 inheriting pht10- ywbn and pht10-gfp were grown in LB medium and treated with 1 mm IPTG. Aliquots were taken before addition of IPTG (t = 0) and 2 and 4 h after addition (t = 2 and 4) OD578 were lysed and analyzed by SDS-PAGE (A) and by a Western blot of the gel was probed with α-cmyc (B), α-htpg as a loading control (C). A B C M h Next, the presence of the c-myc tag was analyzed by Western blot using monoclonal antibodies. In each case, one band became visible with the expected molecular mass (Fig. 3.25B); HtpG served as a loading control (Fig. 3.25C). In summary, the vector plasmid 76

85 pht10 allows overproduction of c-myc-tagged recombinant proteins and their easy detection by immunoblotting Addition of the His- and Strep-tags To improve the versatility of the expression vector by allowing single-step purification of recombinant proteins, two different purification tags were decided, namely the His- and the Strep-tag. While the His-tag is widely used in conjunction with metal chelate resins [64], the Strep-tag has been developed as an alternative tool [139, 140]. The Strep-tag, an eight-amino acid peptide (sequence: WSHPQFEK), has a high specificity and affinity towards streptavidin. Its sequence was derived by selection from a genetic peptide library [140] and later found to occupy the same pocket of streptavidin where biotin normally becomes complexed [138]. Both tags allow a one-step purification of recombinant proteins using affinity chromatography. The coding region for eight histidine residues was inserted immediately downstream of the Pgrac promoter in such a way that the His-tag is located at the N-terminus of the recombinant protein (pht08). The 8-amino-acid Strep-tag was inserted into pht01 in such a way to allow N- and C-terminal fusions (pht09 and pht24, respectively) Addition of the His-tag The functionality of both vectors was checked by first fusing the coding region of the genes yhcs and srta coding for a putative B. subtilis sortase and the core region of the L. monocytogenes sortase A, respectively [111, 125] to the His purification tag present in pht08. Both plasmids were transformed into B. subtilis 1012, and cells were grown in the absence of IPTG to the mid-log phase. Then, production of the recombinant proteins was induced by addition of IPTG and production of the protein was followed by applying samples on a SDS-PAGE followed by Coomassie blue staining and by Western blotting using anti-his antibodies (Fig. 3.26). In both cases, 2 h of induction turned out to be sufficient for high level of production of the recombinant proteins (Fig. 3.26A and C lane T and 2). Next, both proteins were purified using Ni-NTA columns as described [39] (Fig. 3.26C, lanes E1 and E2). Both proteins could be purified to near homogeneity as demonstrated in Fig. 3.26C. The recombinant proteins were confirmed by Western blot (Fig. 3.26B and D). It was concluded that the expression vector pht08 is suitable for overproduction and single step purification of recombinant proteins with a His-tag at their N-terminus. 77

86 Fig Overproduction and affinity purification of proteins with a His-tag. B. subtilis 1012 carrying pht08-yhcs and pht08-srta were grown in LB medium to mid-log, and production of the recombinant proteins was induced by addition of 1 mm IPTG. (A) Cells were lysed and aliquots were analyzed by SDS-PAGE (lane 1, before induction; lane 2 and 3, after 2 and 4 h induction); (B) Western blot analysis using α-his; (C) Cells were lysed and aliquots were analyzed by SDS-PAGE (lane T, total protein), the cellular extracts were applied to appropriate affinity columns, washed and the bound protein was eluted. E1 and E2 stand for the first and second elution step; and (D) Western blot analysis using α-his Addition of the Strep-tag The coding region for the Strep-tag was inserted into pht01 in such a way to allow its fusion either to the N- or C-terminus of the recombinant protein resulting in pht09 and pht24, respectively. To prove the functionality of both vectors, the coding region of gfp was translationally fused to the Strep-tag resulting in pht09-gfp and pht24-gfp. Both proteins could be produced after IPTG induction though the strain where the Strep-tag was fused to the C-terminus produced higher amounts of GFP (Fig. 3.27). With both tags, GFP could be purified to near homogeneity using an affinity column with immobilized streptavidin. 78

87 Fig Overproduction and affinity purification of proteins with a Streptag. B. subtilis 1012 carrying pht09-gfp and pht24-gfp were grown in LB medium to mid-log, and production of the recombinant proteins was induced by addition of 1 mm IPTG. (A) Cells were lysed and aliquots were analyzed by SDS-PAGE (lane 1, before induction; lane 2 and 3, after 2 and 4 h induction); (B) Western blot analysis using α-strep; (C) α-htpg as loading control. (D) Cells were lysed and aliquots were analyzed by SDS-PAGE (lane T, total protein), the cellular extracts were applied to appropriate affinity columns, washed and the bound protein was eluted. E1 and E2 stand for the first and second elution step; (E) Western blot analysis using α-strep; (F) Eppendorf tubes contain aliquots after elution. 3.4 Establishment of a simple method for identification and screening of strong and efficient promoters To be able to screen for strong promoters easily, a promoter-probe vector has been developed, which uses the bgab-encoded β-galactosidase as reporter enzyme. Furthermore, this vector carries the laco operator, the laci gene encoding the Lac repressor and a MCS allowing insertion of promoters. The laco and laci elements convert any promoter into a controllable promoter, which can be induced by IPTG. Therefore, it should be able to insert even very strong promoters into this new promoter-test vector to be cloned them in the absence of IPTG in E. coli. The following chapters first describe construction of the promoter-probe vector and then analyze their induction in relation of the IPTG concentration added to the medium. 79

88 3.4.1 Construction of the promoter-probe vector pht06 In order to screen for strong promoters in B. subtilis, the promoter-probe vectors placz [176], pbgab [99], pndh05 and pt05z have been used in the past. Plasmid pndh05 is similar to pht06 but does not contain laco and laci and in pt05z bgab was replaced by lacz to screen for promoter strength. However, we failed at the very first step, to clone promoters P88, P89 and some others in E. coli using pndh05 or pt05z (data not shown). The reason for this failure is most probably overproduction of the bgab or lacz encoded enzymes leading to death of the E. coli transformants. To overcome this problem, the laco operator and the laci gene were introduced to control expression of bgab in both E. coli and B. subtilis (2.7.4) resulting in the novel promoter-probe plasmid pht06 (Fig. 2.4). This plasmid was constructed based on the backbone of plasmid pmtlbs72 [156] and its derivative pht01 [105] as described in Material and methods (2.7.4) The use of the promoter-probe vector pht06 for cloning and analysis of strong promoters In order to check whether plasmid pht06 will accept strong promoters, the two synthetic promoters P88 and P89 were introduced. As to be expected, the cloning step in E. coli is rather simple by allowing to screen for blue colonies on X-gal plates in the low concentration of IPTG. Plasmids were then extracted for restriction enzyme analysis, DNA sequencing and transformation into B. subtilis. The BgaB activity of the B. subtilis strains harbouring those plasmids was measured. The activity was higher than that from the strong promoter P grac, pht01-bgab [105], which served as a reference in all the following experiments. Based on this positive finding, more than 80 different promoters were introduced into this plasmid to analyse for the influence of different promoter elements (Fig. 3.29A and part 3.5) and stabilizing elements (Fig. 3.29B and part 3.6). This result indicates that the first problem of cloning and maintaining strong promoters has been successfully solved Linearity of BgaB activity and IPTG concentration Next, it was attempted to measure the activity of different promoters induced at 1 mm and 0.1 mm of IPTG to compare their strength, which did not work properly (data not shown). But while analyzing with the libraries, it was found that there was a correlation between BgaB activity and IPTG concentration. To compare the promoter activities, the linear range between the IPTG concentration and the β-galactosidase activity was investigated, using the three different strong promoters P69, P78 and P223 and P grac as a reference [117]. The results 80

89 revealed a linear range for all promoters with an IPTG concentration from to mm added to the culture medium (Fig. 3.28). Furthermore, it turned out that the weaker promoters had an extended linear range. P69 and P78 could be induced with 0.05 mm of IPTG and P grac with up to 0.1 mm. The data also showed that using high concentrations of IPTG, for example 1 mm for very weak promoters such as P lepa, P85 and P70, better resolution between the different promoters was obtained (data not shown). These results further demonstrate that the second problem to compare the strength of different promoters has been solved. Fig Linear range between IPTG concentration and BgaB activity. The samples were collected 2 h after IPTG addition (0.0025, 0.005, 0.01, 0.025, 0.05, 0.1 and 1 mm). The promoters P grac, P69, P78, P223 were analyzed. The data represent the averages of three independent experiments Observation and measurement of the promoter strength All promoters from the library were analyzed on X-gal plates containing different concentrations of IPTG. Plates were incubated at 37 o C, and pictures were taken by a digital camera (Fig. 3.29). As to be expected, the colour of the colonies could be distinguished at concentrations of IPTG between and mm. Representative colonies on plates with 0.01 or 0.1 mm IPTG are shown in Fig. 3.29C and the activity of promoters from plates with 0.01 mm IPTG is easier to be distinguished based on the colour. Higher concentrations of IPTG did not allow to resolve the strength of strong promoters, but resulted in a better resolution for weak promoters. Based on these results, the concentration of 0.01 mm IPTG was chosen for further investigations. Fig. 3.29A and 3.29B exhibit the activity of different 81

90 promoters on X-gal plate with 0.01 mm IPTG and bluer colour indicates stronger promoters. In addition, the pictures (Fig. 3.29A, 3.29B) were analysed by the QuantityOne programme (Biorad) and the intensity of the blue colour was converted into grey values, which represent the strength of the promoters; higher values represent stronger promoters. From these plate experiments, the strength of the promoters could be visualized based on the intensity of the blue colour of the colonies (Fig. 3.29) and classified based on the grey value of colonies (Fig and data not shown). Fig Screening the strength of strong promoters on LB plates. All clones were spotted on LB X-gal plates containing 0.01 mm IPTG and incubated at 37 o C for 16 h (A and B). Different clones are shown to study the influence of promoter elements (A) and 5 -mrna stabilizing elements (B). Below the plates with the colonies, the numbers of the different promoters are given, which will be described below in detail. (C) Influence of IPTG concentrations (0.01 and 0.1 mm) on the development of blue colour of the colonies; the colonies from (A) indicated by a rectangle were displayed grown in the presence of two different IPTG concentrations as indicated. In addition, all promoters were also evaluated by BgaB measurement 2 and 4 h after addition of IPTG. The strength of the promoters was almost in agreement with the data obtained from the colour of the colonies (Fig and Fig. 3.33B). A major difference between the plate and the liquid assays is that reading of the blue colour on plates occurred during stationary 82

91 A B growth phase, while determination of the β-galactosidase activity used cells in the exponential growth phase. This might explain some of the slight differences. Fig. 3.30A and Fig 3.30B showed a good agreement of grey values and BgaB activities of strong σ A -dependent promoters in B. subtilis. Moreover, the strength of the promoter P223 was 23 times higher than P grac 4 h after addition of 0.01 mm IPTG and P85 was 5 times weaker than P grac. From these data, it can be inferred that choosing the appropriate concentration of IPTG within a linear range is important for a successful analysis. It also turned out that incubation of the plates at 30 o C yielded better results than incubation at 37 o C, and there were no differences between the promoters when the plates were kept from h in the incubator (data not shown). Fig Analysis of promoter strength based on grey values of colonies and their BgaB activity. Samples were arranged in increasing grey values. Labels of partial promoters in A and B here are comparable with A and B in Fig The intensity of colonies from pictures was analyzed by the QuantityOne programme (Biorad). The data shown here were the averages of at least two independent plates or BgaB measurements 2 h after IPTG induction. The two or more values differed among themselves by <15% for samples with activities 10 times higher than those obtained with the P grac promoter and <5% for other samples. 83

92 In summary, these results clearly show that screening of a library of different promoters can be done first on LB plates containing 0.01 mm IPTG and based on this analysis, appropriate colonies can be selected for further analysis. The plate method developed here saves time as it circumvents analysis of each promoter by the time-consuming β-galactosidase measurement. 3.5 Elements of strong σ A -dependent promoters in B. subtilis A total of 40 different synthetic and groe-modified σ A -dependent promoters (Fig 3.31) were introduced into the promoter-probe plasmid, pht06, and analyzed. Strong promoters could be screened for and classified based on the strength of promoters as described before. The further analysis focused on the influence of five different on the promoter strength: (1) the transcriptional start site, (2) the -10 region, (3) the -15 region, (4) the -35 region and (5) the upstream region. BgaB activities and Northern blot analyses were carried out to measure the activity of the different promoters. The analysis of combinations of core promoters and UP elements on gene expression revealed that the β-galactosidase activity expression levels could be increased up to 23-fold and the mrna levels up to 43-fold as compared to the strong P grac promoter, and expression of bgab, under control of these new promoters, could reach up to 30% of the total cellular protein P grac is a strong σ A -dependent promoter for B. subtilis One of the new synthetic strong promoters called P grac was constructed using the B. subtilis promoter groe and the laco operator from E. coli (Fig. 3.32A). Expression of bgab, htpg and pbpe under control of this promoter resulted in 10%, 12% and 16%, respectively, of recombinant protein of the total cellular protein, and this promoter is 30 and 60 times stronger than P xyla and P spac based on the BgaB activity [117]. Northern blot analysis demonstrated that the amount of bgab mrna expressed from P grac, (plasmid pht01 [105]) is 10 and 18 times higher than P spac and P lepa, respectively (Fig. 3.32B). These data demonstrate that P groe is indeed a strong promoter and the question was raised which genetic element(s) is responsible for the high expression level. To be able to dissect and to analyse the different promoter elements separately, the new promoter-probe plasmid pht06 was used that facilitated generation and analysis of promoter libraries. Plasmids with different groe-modified σ A -dependent promoters were generated (Fig 3.31 for promoter sequences). The grey values of transformants, BgaB activities (Fig and 3.30) and the Northern blot analyses (Fig and 3.34) were used to determine the activity of these promoters. 84

93 Core promoters P65 (Pc2mu) P66. (PlacUV*) TAGACAAACTATCGTTTAACATGTTATACTATAATATGCG TTGACACTTTATCTTCCATCTGGTATAATAAATAGAGCGG P67 (PaprE*) TTGACAAATATTATTCCATCTATTACAATAAATTCAGCGG P85 (P lepa ) TTGAATCTTTACAATCCTATTGATATAATCTAAGCTGCGG P70 (P groe ) P57 (-35, -15) TTGAAATTGGAAGGGAGATTCTTTATTATAAGAATTGCGG ----C ATG P58 (-15, -10) ATG----A P71 (-35) ----C P72 (-10) A P73 (-15) ATG P74 (-15) TG P75 (-15) A-G P76 (-15) TTTA-G P77 (-22) TAT P78 (-35, -10) ----C A P79 (-35, -15, -10) ----C ATG----A P80 (-35, -15, -10) ----C TAT----ATG----A P81 (-22, -15) TATTTTA-G P82 (-35, -22) ----C TAT P83 (-35, -25) ----C----AC P87 (-35, -15, -10) ----C TATTTTA-G-----A P88 (-35, -15, -10) ----C-----GTATA----ATG----A A---- P89 (-35, -15, -10) ----C-----TTTAA----ATG----A--GG-T--A Core promoter Upstream Upstream region of core promoter -36 P60 (groe WT ) AAAGGAGGTAAGGATCACTAGAAAATTTTTTATCTTATCAC P59 (groe -42 ) AAAA----- P61 (groe -36 ) G P62 (groe 38 ) CT-- P63 (groe 37 ) T- P64 (groe -37_-42 ) AAAA---T- P70/P80 (HT06) ACCGGAATTAGCTTGGTACCAAAGGAGGTAAGGATCACTAG P95/P90 (groe ) AAAAGAATGATGTAAGCGTGAAAAATTTTTTATCTTACTAG P96/P91 (groe ) AAAA-CT-G P97/P92 (hag) CCTCACAAAAAAAGTGAGGATTTTTTTATTTTTGTACTAGT P98/P93 (lepa) ACCCAGATATGATAGGGAACTTTTCTCTTTCTTGTTACTAG P99/P94 (c2mu) TGTCAACATGAGAATTCTTATCATCAATTTTTGAAAACTAG Full promoter UP P64 AAAATTTTTTAAAAAATCTCTTGAAATTGGAAGGGAGATTCTTTATTATAAGAATT P91 AAAATTTTTTAAAAAACTAGTTGACATTGGAATATAGATATGTTATAATAAGAATT P68 AAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGATATGTTATAATAAGAATT P69 AAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGATTCTTTATAATAAGAATT P100 AAAATTTTTTAAAAAATCTCTTGACATTGGAAGGGAGATATGTTATTATAAGAATT P252 AAAATTTTTTAAAAAATCTCTTGACACTTTACAATCCTAATGATATTATCTAAGCT Fig Sequence of modified promoters based on the promoter of the groe operon (P70). Only chances are indicated within promoters P57-P89 from P70, and within promoters P59-P64 from P60. 85

94 A P groe P grac laco bgab B Relative mrna levels Fig The P grac promoter. (A) Genetic features of the P grac promoter; (B) Northern blot analyses of different promoters as indicated. Relative mrna expression levels were calculated by the QuantityOne programme (Biorad), BgaB activities were measured using 1 mm IPTG Influence of UP elements on gene expression It has been reported that UP elements can enhance the transcription initiation in B. subtilis [21, 96]. To answer the question whether the upstream region of the groe promoter functions as an UP element stimulating the core promoter activity in P grac, different promoters were compared. The BgaB activity of the full-length P groe promoter (UP element and core promoter) in P01 and P229 (changing the T in P01 to a C in P229 at +3) were compared with the P groe core promoter lacking the UP element in P70 and a weak promoter such as P lepa in P85 (Tab. 3.1). The relative BgaB activity shows that the complete promoters P01 and P229 lead to activities 10 times higher than that expressed from promoter P70 without the UP element (Tab. 3.1). These results demonstrate that the UP element of P groe plays an important role in the strength of the P grac promoter. This result further indicates that the expression level is largely dependent on the strength of the promoter. This was also the case for the weak promoter P85 (Tab. 3.1), where the BgaB activity is 5 times lower than that measured with P grac. 86

95 Tab Influence of the upstream region of the core promoter P groe on the BgaB activity. Promoters Relative BgaB activity 0 h 2 h 4 h UP + Core elements Core element P01 (Pgrac) P229 (P01 T+3C) P70 (PgroE) P85 (P lepa ) Next, different upstream regions of P groe and of several other promoters were compared (Tab. 3.2). The UP element mutant C2mu derived from a phage φ29 promoter has been shown experimentally to enhance the promoter activity 1.6-fold as compared to the C2 wild type promoter [96]. The UP element of the hag gene could also be shown to stimulate transcription from both σ D - and σ A -dependent promoters in B. subtilis [21]. These UP elements were fused to the groe core promoter, and the strength of these promoters was compared with those of the lepa and groe promoters. The activity of the upstream region of the core groe promoter (P95) with modifications (because of the introduction of a restriction enzyme site) was comparable with that of C2mu (P99), which indicates that the groe promoter contains a strong UP element. When the DNA sequence TCTT located at -42 to was mutated to AAAA (promoter P96), its activity increased two-fold. When the UP elements of the lepa and hag promoters were fused to the groe core promoter (P97 and P98), the activities turned out to be 4-fold lower (Tab. 3.2, left panel). By fusing the UP elements of the groe and C2mu promoters to the strong core promoter P80 (Tab. 3.2, P90, P91 and P94), the BgaB expression levels of those UP elements from groe (pht90 and pht91) were relevant to C2mu (P94). However, when the UP elements of the lepa and hag promoters were fused to the strong core promoter P80 (Tab. 3.2, P92 and P93), the activities were not much lower as in case of fusing with the weaker core promoter P70 (P97 and P98). 87

96 Tab Influence of the upstream region of core promoters on the BgaB activity. Relative BgaB activity Relative BgaB activity UP element Promoter Promoter 2 h 4 h 2 h 4 h groe wt1 P P (-) P P groe P P groe P P lepa P P hag P P C2mu P P Cells were induced with 0.01 mm IPTG for 2 and 4 h; small superscript numbers indicate modification sites when compared with the wild type upstream sequence of the groe promoter. Data are given for the groe core promoter P70 (left panel) and P80 (right panel) in the absence (-) or presence of different UP elements. P grac was called P01 here; lepa, hag and C2mu, represent upstream sequences of the lepa, hag and the mutated of phage φ29 C2 promoters. To have a closer look on the strong groe UP element, it was modified and fused with the core promoter region of the groe promoter (P70) and the strength of derivative promoters was measured. The UP element was enriched for short A and T tracts based on the consensus sequence (-54-nAnnnnnTnnnAAAAnnnTn-36) [57]. The modified groe promoters effected the amount of mrna produced (Fig. 3.33A), the BgaB activity using the plate assay (Fig. 3.33B) and the BgaB activity after 2 h and 4 h of induction (Fig 3.33B). Shorter UP regions seem not to affect the expression level of BgaB when compared to the promoter P01 and P60 (groe WT2 ) (Fig 3.33B). The amount of mrna produced from P60 was slightly higher as compared to P01 30 min after addition of IPTG (Fig. 3.33A) for unknown reasons. This result was in total agreement with the finding in E. coli that the RNAP holoenzyme initially binds to a DNA region covering bp, extending from 55 to +20 [82]. Single changes at -38, TC CT (P62) and at -37, A T (P63) resulted in reduction of the BgaB expression level. Changes at -36, C G (P61) or a combination of -36 and -38 (P95) reduced the activity by at least a factor of two, while changes at -42, from TCTT AAAA increased the activity when compared P95/P96 (Tab. 3.2) and P60/P59 and P63/P64 (Fig. 3.33). Especially, modification of this promoter (P64) based on the consensus sequence above resulted in two times higher BgaB activity and five times higher mrna levels as compared to the wild-type sequence of P01 and P60 (Fig. 3.33). From this result, one can conclude that the modified nucleotide sequence of the groe UP element, -55-AAAATTTTTTAAAAAATCTC-36 is an active UP element for σ A -dependent promoters, at least in this case of groe promoter. 88

97 A Relative mrna levels B Name of the promoter with a modified UP element Fig Analysis of bgab expression under control of different promoters with modified UP elements. (A) Northern blot analysis; (B) grey values of colonies (partially extracted from Fig. 3.29), and relative BgaB activities. Relative mrna expression levels were calculated by the QuantityOne programme (Biorad) Influence of the +1 region on gene expression To investigate the influence of transcriptional start site region on the expression level, constructs with different nucleotides at the +1, +2 and +3 positions were changed or introduced (Tab. 3.3). The BgaB expression level increased by 20% when changing the T in P01 to an A in P205 at +1, while it produced only a slight increase when combined with the strong promoter P80 resulting in P84. Addition of two additional CC (pht106) or GG (pht86) nucleotides resulted in a more than 3-fold reduction of the BgaB expression level. By changing the U residue at +3 in P205 to G in P104, the BgaB expression level was increased more than twice, most probably by enhancing the mrna stability. 89

98 Tab Influence of the +1 region on the expression of bgab. Original Relative BgaB activity Promoter promoter Sequence 2 h 4 h P01 P01 +1-UGUGG-lacO P205 P01 +1-AGUGG-lacO P104 P01 +1-AGGGG-lacO P106 P01 +1-ACCGCGG-lacO P80 P80 +1-TGCGG-lacO P84 P80 +1-AGCGG-lacO P86 P80 +1-AGGGCGG-lacO Promoters P104, P106 and P205 are different from P01 (P grac ), P84 and P86 are different from P80 as indicated in the sequence. Potential transcriptional start sites are underlined Influence of the core promoter on gene expression levels In this part, the core groe promoter was modified and compared to the wild-type sequence (Fig. 3.31). I also analysed the following published promoters: (i) C2 derived from the lysogenic phage φ29 [96] (in P65); (ii) the E. coli lacuv5 promoter carrying a promoter-up mutation [59] (in P66); and (iii) an apre derivative promoter [160] (in P67). These promoters exhibited a two- to eleven-fold higher BgaB activity as compared to P01 (Tab. 3.4). The modifications of the groe core promoter with changes to reduce G residue within the -23 region (from -26 to -21) resulted in P77, P81, P80 and P82, the BgaB expression levels reduced slightly as compared to the original promoter P70, P76, P79 and P71 (Tab. 3.4) Influence of the -10 and -35 regions on the BgaB activity RNAP from diverse bacteria recognizes the same housekeeping-dependent promoters with the TTGACA (-35) and TATAAT (-10) consensus elements, as defined for the E.coli RNAP, indicating conserved features of this class of eubacterial promoters [57, 101]. To determine WT the influence of the -35 region on the BgaB activity in B. subtilis, the -35 region of P groe TTGAAA (P70) was replaced by the TTGACA consensus sequence (P71). When the BgaB activity of this new promoter P71 was measured, it was increased about 7-fold in B. subtilis and 2-fold in E. coli (Tab. 3.4). In the case of the -10 region, the TATTAT sequence in P70 was changed to the TATAAT consensus one (P72). The activity of BgaB was increased about 10-fold in B. subtilis, but only twice in E. coli (Tab. 3.4). When both consensus sequences were combined in the promoter P78, the amount of transcript was increased 38.8-fold (Fig. 90

99 3.34) and the relative BgaB activity 96-fold (Tab. 3.4). These results are in agreement with the consensus hexanucleotide core elements of strong promoters [57, 101]. Tab BgaB measurement of core promoter regions in B. subtilis and in E. coli. Promoter Relative activity in B. subtilis Relative activity in B. subtilis Promoter 0 h 2 h 4 h 0 h 2 h 4 h P57 (-35, -15) P79 (-35, -15, -10) P58 (-15, -10) P80 (-35, -15, -10) P65 (C2) P82 (-35, -22) P66 (lacuv5) P83 (-35, -25) P67 (apre) P87 (-35, -15, -10) P77 (-22) P88 (-35, -15, -10) P78 (-35, -10) P89 (-35, -15, -10) Relative activity in E. coli P70 (groe WT ) P70 (groe WT ) P71 (-35) P71 (-35) P72 (-10) P72 (-10) P73 (-15) P73 (-15) P74 (-15) P74 (-15) P75 (-15) P75 (-15) P76 (-15) P76 (-15) P81 (-22, -15) P81 (-22, -15) Influence of the -15 region on the BgaB activity in B. subtilis and E. coli Another set of mutations, located in the -15 region (between positions -20 and -13) was generated. It is known that the -15 region ( -20 TTTATGTT -13 ) shown to exert high level of expression in E. coli [89] as in P76 and P81 exhibited only a slight effect in E. coli and none in B. subtilis (P76 and P81; Tab. 3.4). Further studies identified sequence elements, including the dinucleotide TG at position -15 and -14 (designated the extended -10 region), which is the most highly conserved DNA sequence in promoters from gram-positive bacteria but not from E. coli [49, 57, 101]. In the present study, changes at position -15 and -14 of the dinucleotide CT (P70) to TG (P74) had an apparent effect in increasing the activity 15 times when compared to P74/P70 in B. subtilis and not at all in E. coli (Tab. 3.4). This result is in agreement with the dinucleotide TG at position -15 and -14 having a more pronounced effect on transcription in B. subtilis than in E. coli [59]. An A residue at the -16 position seems to be 91

100 more important in increasing the promoter strength in B. subtilis (5-fold increase; compare P73 and P74) than in E. coli (3-fold), ATG at the -16 to -14 region (P73) increased the activity 94-fold as compared to the core groe promoter P70, while moving the ATG region upstream of the -17 to -15 region in P75 resulted in an about 40-fold drop in the BgaB activity (compare P73 with P75; see the DNA sequence in Fig and the relative BgaB activity in Tab. 3.4). These results suggest that promoters with the -15 ATG extended region could enhance the bgab expression more than 94-fold as compared to the wild-type core promoter of groe, and 8-fold higher than P grac. bgab Relative mrna levels Fig Northern blot analysis of bgab expression under control of different promoters. These promoters are derivatives of the wild-type groe promoter and carry either mutations of the core region or are combined with UP elements or both. Relative mrna expression levels were calculated by QuantityOne programme (Biorad) Analysis of combinations of the -10, the -15 and the -35 regions The combination of the -35 or the -10 consensus sequence with the -15 ATG region resulted in promoters P57 and P58 (see Fig for sequences), and both of them were analysed by Northern blot (Fig. 3.34) and measurement of the BgaB activity (Tab. 3.4). The results revealed a 38- to 42-fold higher mrna level when they were compared with P grac (Fig. 3.34), and 12-fold higher BgaB activity (Tab. 3.4). The promoter with a combination of both the -35 and -10 consensus sequence with the -15 ATG region was named P79, and some derivatives called P80, P87, P88 and P89 (sequences in Fig. 3.31). When their relative BgaB activities WT were measured, they were 150 to 180 times higher as compared to P groe (P70) (Tab. 3.4). Clearly, the strong groe promoters include not only the TTGACA (-35) and TATAAT (-10) hexanucleotide core elements but also an ATG trinucleotide at position -16, -15,

101 3.5.5 Combinations of UP elements and core promoters The previous results already showed that a modified UP element of the groe promoter, (P64) enhanced expression of bgab. Will this improved UP element further increase the activity of already strong promoters? To answer this question experimentally, this strong UP element was combined with the core promoter regions from P79, P78 and P57 resulting in P68, P69 and P100. The BgaB activities of these promoters are presented in Tab. 3.5, the amount of BgaB protein in Fig. 3.35, of the transcripts in Fig The new promoters did not significantly enhance bgab expression when compared with P57, P78 and P79. This result indicates that a combination of a strong UP element and a strong core promoter does not further increase expression. A core promoter itself is sufficient to obtain high gene expression levels. However, the influence of the two new promoters P68 and P100 with the -15 ATG region revealed that the β-galactosidase activity levels could be increased up to 13-fold and the mrna levels up to 43-fold as compared to the strong P grac promoter. This corresponds to an about 690-fold increase when compared with the well-known P spac promoter, and expression of bgab under control of these promoters reached up to 30% of the total cellular protein (Fig. 3.35, P68, P100). I also checked for the production of another intracellular recombinant proteins (Pbp4*). Expression of pbpe under control of the promoter P100 reached up to 38% of the total cellular protein (Fig. 3.43). Those data also corroborated previous observations that the -15 region plays an important role in increasing the strength of the groe promoter. Tab BgaB measurement of consensus promoters in B. subtilis. Original core Relative BgaB activity Promoter promoter 0 h 2 h 4 h P01 (grac-full) P groe P68 (UP*, -35, -15, -10) P groe (P79) P69 (UP*, -35, -10) P groe (P78) P100 (UP*, -35, -15) P groe (P57) P252 (UP*, -35, -15, -10) P lepa (P85) UP*: UP element from P64 was used for this experiment. 93

102 Fig Production of BgaB from different promoters. Different concentrations of IPTG were applied as indicated, and proteins were separated by 10% SDS-PAGE. The percentage of the expressed BgaB protein was calculated by AlphaEaseFC (Alpha Innotech) programme. In case of the weak promoter lepa, even though the core promoter region of the lepa promoter (Fig. 3.31, P85) has a good -15 and -10 region ( -15 TGATATAAT -7 ), the expression level is rather low when compared with the corresponding modified P groe (Tab. 3.4, P72 and P74). Promoter P252 was constructed by a combination of a strong UP element, the -35 and -15 region of P100 and the core lepa promoter. This new promoter improved the lepa promoter about 5-fold when compared to P85 (Tab. 3.5) at 0.01 mm IPTG, and the mrna levels increased up to 2.4-fold as compared to the strong P grac promoter (Fig. 3.34). However, when the BgaB activity induced with 1 mm IPTG was determined with P252, it turned out to be 60 times higher than that expressed from P85 and P lepa wild-type (data not shown) as in the case of P grac (Fig. 3.32). 3.6 Using stabilizing elements to enhance the protein expression level in B. subtilis In order to improve the productivity of B. subtilis, efforts have focused on the development of protease deficient strains [170] and the identification of efficient regulatory elements at the transcriptional (see in Results part at 3.5), the translational and the protein secretion levels [172], but the potentiality of transcript stabilization has hardly been investigated [32]. Decay of mrna plays an important role in gene expression control. The main factors affecting mrna stability in bacteria are translation initiation frequency, codon usage and RNA secondary structures. Efficient translation initiation and consequent immediate ribosomal protection from degradation stabilizes the mrna and is achieved by selection of ribosomal 94

103 binding sites lacking inhibitory secondary structural elements. Stable hybrid mrnas might be constructed by implementation of efficient 5 and 3 stabilizing sequences as a barrier against exonucleases [124, 150]. In this work, mrna stabilizing elements were also analyzed by using a similar experimental approach. First, 17 different 3 -mrna terminal stem-loops have been investigated. Second, the 5 -mrna stabilizing elements including a strong RBS, the laco controllable stabilizing element (CoSE) and the spacer between RBS and CoSE were examined (Fig. 3.36). The results demonstrated that the CoSE together with an appropriate spacer and a strong RBS could increase the protein expression level 9-fold as compared to the P grac promoter, resulting in up to 26% of the total cellular protein and a half-life of the mrna of more than 60 min. A combination of strong promoters and stabilizing elements revealed that recombinant protein expression levels up to 42% of the total cellular protein could be obtained. P grac P groe - TGTGG...lacO...CCcaatt...RBS Stem-loop CoSE Fig Genetic features of the P grac promoter. Spacer mrna terminal stem-loops Protection of the 3 end has been less well studied for B. subtilis than for E. coli. E. coli transcription terminators and other stable secondary structures inhibit the progress of both PNPase and RNase II. The cryiaa transcription terminator serves as a stabilizer of exonucleases heterologous RNAs in B. subtilis [169], suggesting that this phenomenon may be shared by both organisms even though PNPase is the only major exonuclease they have in common [28]. To further examine the specificity of the enhancement effect of the trpa terminator fragment at the 3 end, plasmid pht36 was constructed by removing the trpa transcriptional terminator from pht01-bgab (see in 2.7.6, Fig. 2.5 for genetic features, and Fig 3.37 for the pedigree of plasmids). Then, terminal stabilizing elements were introduced into pht36 at the AatII restriction site, and the cells were cultured for the measurement of β-galactosidase activity (Tab. 3.6). When the OD 578 reached to 0.8, 0.1 mm IPTG was added and cells further incubated for 30 min. Then, rifampicin was added and samples were taken 95

104 just before addition of rifampicin (0) and 5, 10, 15, 20 30, 45 and 60 min thereafter. Total RNA was prepared, separated through an 1.2% agarose gel, blotted and probed with antibgab RNA. Densitometric quantification was performed using the QuantityOne programme. The mrna half-life was calculated by setting the control value (t = 0) to 100% (Fig and Tab. 3.6). Fig Relation between the plasmids pndh33, pht01, pht06 and pht36. -, removal of the repeated 117-bp region; P grac promoter or trpa terminator; +, addition of bgab gene or MCS. 96

105 Tab Sequences of 3 -mrna terminal stem-loops and their influence on mrna stability and BgaB activity in B. subtilis. ΔG Relative BgaB activity Name Origine Sequence of the 3 -mrna terminal stem-loops kcal/mol 2 h 4 h S01 trpa WT GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUUUUUUC S36 orf CAACAAAGGCUGAGACAGACUCCAAACGAGUCUGUUUUUUUAA S110 skfa GACGUCUUUGAGAAUAGGGAGUUGAGCGUAUUUGCUUAUACUCCUUAUUUUCUUUUUUUU S111 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCUUUUUC S112 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCCUUUUUC S113 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCCCUUUUUC S114 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCCCCUUUUUC S115 trpa WT GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUUUUUUC S116 ** GACGUGAAAAAAGCCCGCUCAUUAGGCGGGCUGCCCCGGGGACGUC S117 * GACGUCCCCGGGCAGCCCGCCUAAUGAGCGGGCUGCCCCUUUUUC S118 * GACGUCCCCGGGGCAGCCUGCCUAAUGAGCGGGCUGCCCCUUUUUC S119 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCCCUUUC S120 * GACGUCCCCGGGGCAGCCCGCCUAAUGAGCGGGCUGCCCUUC S121 * GACGUCUAGCCCGGGUAAGCUUCGGCCGAGCCCGCCUAAUGAGCGGGCUCGGCCUUUUUC S122 * GACGUCUAGCCCGGGUAAGCUUCGGCCGAGCCCGCCUAAUGAGCGGGCUCGGCCUUUC S123 htpg GACGUCAAAAAGGAAUCGUCUCAUAAGGAGACGAUUCCUUUUUUUC S124 dnak GACGUCAGAAAGUCAAAGUCAGGCAUCUCUUGGCUUUGACUUUUUUUC S36 were generated by removal of trpa termininator from S01 (plasmid pht01) and the terminator of the transcriptional unit is at the end of orf-3 on plasmid pht01. All stabilizing elements (terminators), from S110 to S124 were introduced into plasmid pht36 at AatII site (underline). trpa WT, wild type E. coli trpa terminator; *, modification of E. coli trpa terminator; **, reversed complement of S115. skfa, stem-loop between B. subtilis skfa and skfb with additional of poly T. htpg and dnak are terminators of B. subtilis htpg and dnak genes. The free energy was calculated by RNAfold programme at 37 o C. 97

106 Fig Rifampicin experiment for determination of mrna stability. The plasmid pht36 carries different transcription terminators. The cells were cultured in LB medium at 37 o C, and when an OD 578 of 0.8 was reached, 0.1 mm IPTG was added and the incubation continued for 30 min. Then rifampicin (100 μg/ml final concentration) was added. Total RNA was prepared at the time points indicated and probed with anti-bgab RNA. Densitometric quantification was performed using the QuantityOne programme (Biorad). The mrna half-life was calculated by setting the value of control (t=0) to 100%. The marked differences between the plasmid pht01-bgab (S01) and pht36 (S36) presented in Fig. 2.5, in which S01 contains the E. coli trpa terminator and a truncated orf-4 (function unknown) [156], while these sequences are absent from S36. The expression level from S36 was slightly lower than that from S01 (Tab. 3.6 and Fig. 3.38). S01 and S115 use the same E. coli trpa terminator, but S115 does not contain the truncated orf-4. The expression levels of these two constructs were comparable. These results indicate that the E. coli trpa terminator slightly influenced both the BgaB activity and mrna stability. The data in Tab. 3.6 indicate that lowering the free energy in S113 and S121 resulted in a slight reduction of the BgaB activity. However, after removal of some T-residues (Tab. 3.6; S119, S120 and S122), the BgaB activity increased. It follows that leaky transcription of the 98

107 downstream genes, orf-3 (function unknown) and orf-1 (Fig. 2.5), will increase BgaB expression. This phenomenon might explain the result of leaky expression of orf-1, which is supposed to code for RepA, and by increasing the amount of RepA the copy number of the plasmid is increased, too [156, 157]. The orf-1 product (RepA) displays homology to DnaA chromosomal initiators and carries a DNA binding (HTH) motif and the intergenic region encompasses repeated sequences (iterons) and a dnaa box [156, 157]. It has been proposed that binding of RepA to the iterons is a prerequisite for initiating plasmid replication [33]. One might exploit this characteristic of the plasmid to construct a copy-number control regulated expression plasmid for B. subtilis like the E. coli pbac/oriv vector [165]. In general, the expression levels of bgab with differently modified trpa stem-loop terminators or some natural stem-loop terminators of skfa, htpg and dnak were similar. Decreasing or increasing the free energy of stem-loop structures did not increase significantly the BgaB activity (30% increase or decrease), and the average half-life of the bgab transcripts is 30 min, similar to S01 from pht01-bgab. In summary, data from these experiments do not recommend using these 3 -mrna terminal stem-loop structures to stabilize mrnas for enhancing protein production The 5 stem-loop structure (laco stabilizing element) It is known that the 5 -end of transcript plays an important role as a stabilizing element [15, 16, 38, 90, 146]. It was asked whether the laco operator can be used to both to control expression of recombinant genes and to act as a 5 stabilizing element. The laco of P grac (Fig. 3.36) was used for further modifications. The different promoters with modification of the laco stem-loop structure were inserted into the promoter-probe plasmid pht06. To increase the free energy (ΔG) of the laco stem-loop structure, I modified some nucleotides outside of the laco terminus resulting in S102 (ΔG = -6.3) and S103 (ΔG = -2.8) (S01 from P grac ; ΔG = -9.8) (Fig and Tab. 3.7). The results showed that a higher free energy of the stem-loops in S102 and S103 lowered the BgaB activity and reduced the mrna stability (Tab. 3.7 and Fig. 3.39). As mentioned before, a high GC at the +1 region decreased the BgaB activity (Fig. 3.9 and Tab. 3.7, S106, S107, S108 and S109) due to reduction of the amount of mrna molecules (data not shown), though the mrna stability of S109 increased up to 60 min. Keeping ΔG around in the cases of S201, S205, S228 and S229 did not change the BgaB expression level significantly. In contrast, the bgab mrna stability was decreased to 13.9 min (S201) and 18.7 min (S205), and kept at about the same level in the case of S1 (

108 min), S228 (26 min) and S229 (30 min) (Fig.3.39, Tab. 3.7). This result indicates that there might be interferences between the laco stem-loop and the stabilizing RBS. Based on published data [146], a decrease in free energy of the stem-loop at the 5 -terminus should increase the mrna stability. Addition of some nucleotides outside laco, S104 (-12.9), S105 (-16.4), S203 (-14.1), S207 (-16.9), S208 (ΔG = -16.9) and S209 (-16.1) turned out to increase the mrna stability from 32 to 60 min. The stabilizing elements S104, S105, S203, S209 could increase the BgaB activity 1.5- to 2.8-fold as compared to S01, while S207 and S208 exerted the same influence as S01 (Tab. 3.7). Unexpectedly, S202 (ΔG = -14.4) reduced both the mrna (21 min) and the BgaB (0.7) expression level (Tab. 3.7, S202). This result indicates that increased mrna stability is not always accompanied by an increase in the BgaB expression level, which reinforces the assumption that there is an influence between the stem-loop and the RBS leading to a lower BgaB expression level. On the other hand, analysis of modified nucleotides inside laco to decrease the free energy of the stem-loop as seen with S206 (ΔG = -13.7) and S211 (ΔG = -11) did not increase the mrna stability, but the BgaB activity was enhanced more than 2-fold as compared to the P grac promoter (Tab. 3.7). The modification inside laco especially draw my attention because of a lower background level (data not shown) as compared to the other sequences, and it will be easy to decrease the free energy of the stem-loop by addition of other nucleotides outside laco for further study of protein expression. It can be concluded that a lower free energy of the laco stem-loop at the 5 -terminus of transcripts resulted in a higher mrna stability, even there were some exceptions, but the BgaB activity is still not very high. This result suggests a putative interference between the stem-loop structure and the stabilizing RBS element. I decided to investigate the interference between the strong RBS and the spacer region between the laco stem-loop and the RBS. Generally, the promoters S104, S105, S206, S209 and S211 could be used for further investigations. However, only the promoters S206 and S211 have been used for further modifications of laco because they revealed a lower background and the possibility to decrease the free energy below -16.6, while avoiding too many GCs at the transcriptional start site region. 100

109 Tab Sequences of 5 -mrna stabilizing elements and their influence on mrna stability and BgaB activity in B. subtilis Name Half-life of Relative BgaB activity ΔG mrna kcal/mol 2 h 4 h (min) Stem-loop sequence Spacer and RBS S UGUGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S UGUGGAAUUGUGAGCGGAUAACAAUUC-----AAUUAAAGGAGG S UGUGGAAUUGUGAGCGGAUAACAAUU------AAUUAAAGGAGG S104* AGGGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S AGGGGAAUUGUGAGCGGAUAACAAUUCCCC--AAUUAAAGGAGG S UGUGGAAUUGUGAGCGGAUAACAAUUCCA---AAUUAAAGGAGG S UGUGGAAUUGUGAGCGGAUAACAAUUCCACA-AAUUAAAGGAGG S AGUGGAAUUGUGAGCGGAUAACAAUUCCACA-AAUUAAAGGAGG S AGUGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S AGUGGAAUUGUGAGCGGAUCACAAUUCCC---AAUUAAAGGAGG S UGGGGAAUUGUGAGCGGAUAACAAUUCCCC--AAUUAAAGGAGG S UGGGGAAUUGUGAGCGGAUAACAAUUCCCC--AACUAAAGGAGG S AGGGGAAUUGUGAGCGGAUAACAAUUCCCC--AACUAAAGGAGG S UGUGGΔAUUGUGAGCGGAUCACAAUUCCC---AAUUAAAGGAGG S228* AGCGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S229* UGCGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S106* ACCGCGGAAUUGUGAGCGGAUAACAAUUCCC---AAUUAAAGGAGG S ACCGCGGAAUUGUGAGCGGAUAACAAUUCCGC--AAUUAAAGGAGG S UCCGCGGAAUUGUGAGCGGAUAACAAUUCCGCG-AAUUAAAGGAGG S ACCGCGGAAUUGUGAGCGGAUAACAAUUCCGCGGAAUUAAAGGAGG All stabilizing elements were fused with the promoter groe. The free energy was calculated by the RNAfold programme at 37 o C. Relative BgaB activities were compared with those measured with pht01-bgab which was set as 1. Densitometric quantifications were performed using the QuantityOne programme. The mrna half-life was calculated by setting the control value (t=0) to 100%. *, Construction of plasmid is described in

110 Fig Rifampicin experiment for determination of mrna stability. All laco variants have been inserted into plasmid pht06. The cells were grown and treated as described in the legend to Fig Influence of a strong RBS It has been reported that a strong RBS could function as an mrna stabilizing element and enhance the protein expression level [7, 8, 32, 44, 45, 53, 54, 74, 134, 145, 146]. The strong B. subtilis gsib RBS [74] was evaluated again in this work. The promoter groe and lepa were fused with the strong gsib RBS, then the recombinant promoters were introduced to pht06, in which laco was removed. I was unable to construct promoters S219 and S220 (Tab. 3.8) and other strong promoters (not shown), indicating that P groe was already a very strong promoter in E. coli, which causes lethal overproduction of BgaB when combined with a strong RBS. This result also indicates that when using stabilizing elements (RBS and 5 -stemloop structures), a strong promoter must be controlled to facilitate cloning in E. coli and expression later in B. subtilis. This could explain the failure to use known stabilizing elements 102

111 for overproduction of recombinant protein. To study controllable stabilizing elements for B. subtilis including strong RBSs for overproduction of proteins is the main task of part When the wild type RBS of lepa was replaced by the strong gsib RBS, the BgaB activity was more than 100-fold higher, and the BgaB protein reached about 2.8% of the total cellular protein (Fig. 3.42), while it was undetectable when synthesized from the lepa RBS (data not shown). In addition, replacement of the RBS increased the half-life of the bgab transcript from 6 to 17 min (Fig. 3.41). These findings are in agreement with the observation that the stability of the gsib mrna depends on the strong RBS [74]. It further indicates that the gsib RBS can be used to enhance the stability of the bgab transcript significantly and most probably of any transcript. Tab BgaB activities and half-lifes of bgab mrna in B. subtilis equipped with different RBS sequences. Relative BgaB activity Half-life of Name Original promoter bgab mrna 2 h 4 h (min) S219 P groe S220 P groe lepa (S02) From phcmc02 < 0.02 < S221 P lepa Name RBS Sequences from transcriptional start site to RBS S219 (P groe ) groe WT RBS TGTGACTATTGAGGAGGTTGGATCC S220 (P groe ) gsib RBS TGTGAAATTAAAGGAGGAAGGATCC lepa (S02) (P lepa ) lepa WT RBS TAGTGTATTTTGCGTTTAATAGTAGGAGTGAGGATCC S221 (P lepa ) gsib RBS TAGTGTATTTTGCGTTTAATTAAAGGAGGAAGGATCC The sequence S02 corresponds to the wild-type lepa promoter-rbs of phcmc02 [104]. This table gives the names of the promoters and of the RBSs used. Potential transcriptional start sites and RBS are underlined. B. subtilis cells were induced with 1 mm IPTG Influence of the spacer length between the laco stem-loop and the RBS From the results shown in 3.6.2, and 3.6.3, it can be concluded that the spacer length between the laco operator and the RBS is an important factor, which contributes to mrna stability and overexpression of recombinant proteins. To screen for the optimal spacer, two DNA sequence sets were analysed one based on S202 (S214 and S215) and the other on S211 (S

112 and S210) using a laco sequence with a free energy ΔG of about Interestingly, the spacer did not only influence the mrna stability, but the bgab expression level (Tab. 3.9 and Fig. 3.40). Spacer lengths of 13, 19 and 29 nucleotides were tested to optimize BgaB expression. A spacer of 19 nucleotides (S213) seems to be optimal for the production of BgaB, in that the BgaB activity could be increased up to 9-fold as compared to the original spacer and the half-life of the mrna was enhanced to more than 60 min (Fig. 3.40). The two promoters P212 and P01 are identical, but differ in their stabilizing elements including the stem-loop and the spacer region. With P212, the BagB protein accumulated to about 30% of the total cellular protein as compared to about 10% with P01 (Fig 3.42). These data indicate that spacers of 13 to 29 nucleotides located between laco and RBS greatly enhanced the stability of the mrna and the protein expression level. Tab Sequence of stabilizing elements with different spacer lengths. Name ΔG Sequence of controllable stabilizing elements S S S S S UGUGGAAUUGUGAGCGGAUAACAAUUCCACAaauuAAAGGAGG (Spacer 4) UGUGGAAUUGUGAGCGGAUAACAAUUCCACAaccaacaccaauuAAAGG AGG (Spacer 13) UGUGGAAUUGUGAGCGGAUAACAAUUCCACAacaacaaccaacaccaau uaaaggag (Spacer 19) UGUGGΔAUUGUGAGCGGAUCACAAUUCCACAaccaacaccaauuAAAGG AGG (Spacer 13) UGUGGΔAUUGUGAGCGGAUCACAAUUCCACAacaacaaccaacaccaau uaaaggag (Spacer 19) S UGUGGΔAUUGUGAGCGGAUCACAAUUCCACAacaacuagauucuaucaa uucaaccaauuaaaggagg (Spacer 29) S202 was used to construct S215 and S215; S211 was used to construct S212, S213 and S Combinations of strong stabilizing elements with different promoters The strong controllable stabilizing elements (CoSE) S212 and S213 were fused to the two weak promoters P lepa (resulting in P224) and P spac (P222) and the two strong promoters P69 (P223 and P225) and P100 (P250) (see description in Tab. 3.10). To assay for functionality of CoSE, the combinations of CoSEs and promoters were inserted into the promoter-probe plasmid pht06. Next, these recombinant plasmids were transformed into B. subtilis 1012, and the BgaB activities were measured (Tab. 3.10). While the activity was extremely low in 104

113 the absence of CoSE, it increased 45-fold when P lepa and P spac were combined with CoSE S212 (P224 and P222) as compared to P lepa (in phcmc02) and P spac (in phcmc05). The half-life of the mrnas also increased (Tab and Fig. 3.41, P lepa /P224 and P spac /P222). The expression level of the BgaB protein under control of P224 (P lepa -CoSE) reached up 7% of the total cellular protein (with did not result in the appearance of any additional band with a molecular mass of BgaB, in the case of P lepa ) (Fig. 3.42, S224). Half-life (min) BgaB 2 h Fig Rifampicin experiment for determination of mrna stability and relative BgaB activity. The plasmid pht06 carried the different promoters with modifications of the spacer length between laco and RBS. The cells were grown and treated as described in the legend to Fig The products expressed from the three recombinant promoters P223, P225 and P250 were analysed. The BgaB activity was enhanced 2- to 3-fold as compared to the original promoters (P68 and P100) and to a more than 23-fold level as compared to strong promoter P grac (Tab. 3.10). I also looked directly for the production of the BgaB reporter protein using the different recombinant promoters P223 and P250, and compared them to P01 (P WT grac ) and P212 (P WT groe -CoSE). As shown in Fig. 3.42, P250 led to the production of BgaB up to 35.5% of the total cellular protein, and P223 could even overproduce BgaB up to 42% of the total cellular protein. While BgaB protein produced from the strong P grac in P01 produced 9.2%, P WT groe -CoSE (P212) produced 26.7%, P68 and P100 (see in Fig 3.35) 30% of the total cellular protein. These results strongly indicate that CoSE could help to enhance significantly both mrna stability and the protein level. 105

114 Tab Combination of stabilizing elements with different promoters (P spac, P lepa, P69 and P100). Relative BgaB activity Promoter Description Half-life 2 h 4 h P01 P grac in pht01-bgab P lepa From phcmc02 < 0.02 < P224 P lepa with S > 60 P spac From phcmc05 < 0.02 < P222 P spac with S P69 P69 (UP*, -35, -10) P223 P69 with S P225 P69 with S P100 P100 (UP*, -35, -15) P250 P100 with S > 60 UP*: The UP element was modified from P64; -35, -10, -15, and the core promoter was modified at the -35, -15 or/and -10 region. Fig Rifampicin experiment for the determination of mrna stability. The plasmid pht06 carried different promoters. The cells were grown and treated as described in the legend to Fig

115 Fig SDS-PAGE of some plasmids with stabilizing elements or recombinant promoters. Different concentrations of IPTG were applied as indicated. The percentage of the produced BgaB in total cellular protein was calculated by the AlphaEaseFC programme (Alpha Innotech). To analyse expression of two additional genes, pbpe and htpg were fused to the strong promoter P250, carrying the stabilizing element S250. Gene pbpe was additionally fused to P100 to allow a comparison. Expression from these promoters result in up to 38%, 28% and 26% recombinant protein of total cellular protein (Fig. 3.43). This result indicates that those promoters and stabilizing elements can be used not only for BagB, but also for other proteins. 107

116 Fig SDS-PAGE of pbpe and htpg fused with promoters P100 and P250. Different concentrations of IPTG were applied as indicated. The percentage of the expressed recombinant proteins Pbp4* and HtpG was calculated by the AlphaEaseFC programme (Alpha Innotech). 108

117 4. Discussion 4.1 Riboswitches in expression systems Expression systems based on strong and tightly regulated promoters play an important role in modern biotechnology (reviewed recently in [97, 163]). The first inducible system for B. subtilis was described by Yansura and Henner [173] based on the artificial spac promoter which is under the negative control of the LacI repressor. Subsequently, induction systems have been constructed based on sucrose [178], phosphate-starvation [87], T7 polymerase [29], xylose [79], stationary phase [70] and citrate [40] as inducer. From all of these different systems, only the IPTG- and the xylose-inducible ones are widely used. The advantages and disadvantages of these different expression systems would be discussed elsewhere [144]. New alternative systems allowing both intra- and extracellular expression are being searched for, which should be easy to use, very efficient and inexpensive. Two of them use the amino acids L-glycine and L-lysine as inducer The glycine system It has been suggested that about 2% of the B. subtilis genes are regulated via riboswitches [93]. If B. subtilis cells grow in the presence of a high L-glycine concentration, they activate expression of the gcv operon whose enzymes degrade L-glycine to 5-10-methylenetetrahydrofolate, ammonia and CO 2 [78]. Activation of the gcv operon occurs by L-glycine itself, which binds to a tandem riboswitch as first suggested and then experimentally proven [5, 93]. Based on this mechanism, we asked whether it can be converted into an expressionsecretion system for B. subtilis and related species. In a first experiment, Northern blot was used to prove that indeed only a short transcript, the attenuation product, was synthesized in the absence of added L-glycine corresponding to the 5' UTR. Upon increasing the L-glycine concentration, full-length transcripts were present already after 5 min. These data fully confirm the in vitro data of Mandal and coworkers [93] that a low L-glycine concentration leads to transcription attenuation due to the formation of a rho-independent terminator. Most interestingly, the attenuation product did not disappear after L-glycine induction. An extended half-life of this short transcript could be excluded by the rifampicin experiment, as well as its production by processing. But processing indeed occurs, and seems to increase the stability of the 3' terminal product coding for the three enzymes. The question remains what might be the function of the riboswitch RNA under inducing conditions? Two possibilities were envisaged, not mutually exclusive. First, the riboswitch RNA could be translated. Indeed, a reasonable 109

118 Shine-Dalgarno sequence can be deduced from its primary sequence (AAGGAG) followed by an open reading frame encoding a potential 25-residues peptide. If this peptide is really synthesized, it might exert a regulatory role as described, e.g., for phage λ CIII and B. subtilis SpoVM peptides, which both inhibit the proteolytic activity of the ATP-dependent metalloprotease FtsH [31, 60]. Another example are the pentapeptides encoded by the phr genes, which inhibit their cognate phosphatases [115]. Second, the riboswitch RNA could act as a nc (non-coding) RNA. More than 60 ncrnas have been described in E. coli, where most of them exert a regulatory function [46, 151]. These results suggest that the riboswitch regulatory region can be used to express recombinant genes. To evaluate its potential, the lacz reporter gene was fused to the riboswitch. The lacz gene offers a simple and fast way to measure promoter strength. First I investigated which L-glycine concentration yielded full induction. It turned out that mm L-glycine resulted in an up to 6-fold increase in the β-galactosidase activity, and all subsequent experiments have been carried out using 10 mm of L-glycine. If this induction factor is compared to those obtained with the most used systems, an induction factor of about 40 was measured with xylose and an about 15-fold induction with IPTG, respectively [55]. Minimal medium yielded a higher induction factor than LB medium due to a higher basal level caused by the unknown L-glycine concentration present in this complex medium. Next, the promoter strength should be increased, and this was obtained when the wild-type -35 element was replaced by its consensus sequence. We noticed that both the basal and the induced level were increased. A further improvement of the -10 region was not possible suggesting that this element is already optimal. Experiments are in progress to reduce the basal level by (i) increasing the number of U residues immediately downstream of the terminator structure (in the wild-type sequence, five U residues are present) and (ii) to increase the strength of the terminator structure to obtain a more efficient transcription termination in the absence of added L-glycine (Fig. 4.1). The U residues immediately downstream of the terminator weaken the interaction between the transcript and the RNAP, and three to seven U residues have been described with natural terminators. We asked, whether the stability of the terminator structure could be influenced by the growth temperature, which in turn would influence the basal level of expression. This turned out to be the case, but only at 45 C; no significant difference could be measured between the two cultures grown at 30 C and 37 C, two temperatures which are normally used for the production of recombinant proteins. 110

119 Fig The glycine-responsive riboswitch. (A) Secondary structure [93]; (B) primary structure; (C) primary structure with a potential open reading frame. The production of two recombinant proteins was also checked, one present in the cytoplasm (HtpG and Pbp4*) and the other secreted into the medium (α-amylase). In both cases, 111

120 detectable amounts of the recombinant protein on Coomassie blue stained SDS-PAGE were only present when the genes were fused to the consensus promoter. While the medium did not significantly influence the amount of HtpG and Pbp4* protein produced, higher amounts of α-amylase were secreted into the medium when the cells were grown in LB medium indicating that less protein is secreted from cells grown in minimal medium. In conclusion, these experiments demonstrate that the glycine tandem riboswitch can be used to obtain regulatable expression of recombinant proteins, both intra- and extracellularly. These data have already been published [116] The lysine system The lysine-responsive lysc gene as a second example for a riboswitch regulated gene was studied, which becomes induced at low L-lysine concentrations within the cytoplasm [51, 153]. Northern blot experiments were carried out to study the switch in transcription of lysc upon removal of L-lysine from the medium. The transcription pattern changed as to be expected from the short riboswitch RNA to the full-length transcript (Fig. 3.15A). Surprisingly, the riboswitch RNA not only continued to be synthesized for at least 3 h, but remained predominant. We could show convincingly that the riboswitch RNA does neither result from increased stability nor from processing of the full-length transcript. Using in vitro single-round transcription with E. coli RNAP in the absence of L-lysine, the riboswitch RNA could be detected, but it corresponded to only ~36% of the total transcript yield [51, 153]. Under in vivo conditions, the riboswitch RNA accumulated to more than 90% of the total lysc-specific transcript (Fig. 3.15A). When the right arm of the transcriptional terminator was removed, the riboswitch RNA failed to be synthesized as to be expected. When the β-galactosidase activity was measured, it turned out to be inducible in the presence of the wild-type terminator (Fig. 3.19), but constitutive when the truncated terminator was present (Fig. 3.20). This finding is surprising since all the sequence information to fold the antiterminator is still present. This result indicates that formation of the active riboswitch able to bind L-lysine needs the sequence information of the complete transcriptional terminator. But these results cannot exclude the possibility that the coding region for lacz interferes with the ability to fold the anti-terminator structure. A detailed mutational analysis has to be carried out to identify those sequences needed to build the active riboswitch. Three important questions arise. By which mechanism continued synthesis of the riboswitch is ensured? What is the biological function of the riboswitch RNA during expression of lysc? 112

121 Why the level of the full-length mrna is very low after removal of L-lysine? Theoretically, the riboswitch RNA could serve as a reservoir for L-lysine. This possibility is rather unlikely because of two reasons. First, the amount of riboswitch RNA molecules though not quantified should be too low to act as a storage devise for L-lysine. Second, the short half-life strongly argues against the storage hypothesis. Based on the finding that the amount of riboswitch RNA remains more or less constant, independent of the L-lyine concentration, one can ask whether it could be translated. A close inspection of the RNA sequence revealed a reasonable Shine-Dalgarno sequence (AGAAAGatGG) and a downstream GTG start codon followed by 23 sense codons as already suggested for the glycine riboswitch. Alternatively, the riboswitch RNA could serve as a ncrna [46, 151] influencing translation of one or more transcripts (Fig. 4.2). Further experiments have to be carried out to test these possibilities. If the riboswitch RNA exerts a function within the cell, which has never been reported so far, low L-lysine concentrations will allow for transcription attenuation only for 5-10% of the riboswitch RNA molecules based on the results shown in Fig. 3.19A, and no specific mechanism will be needed. Alternatively, through less likely, an unknown metabolite could stabilize a secondary structure different from the one recognized by L-lysine thereby favouring transcription attenuation. Furthermore, B. subtilis produces three aspartate kinases, I, II, and III, regulated by different end products; the gene dapg coding for aspartate kinase I, the gene lysc coding for aspartate kinase II are regulated by diaminopimelate and lysine, respectively, and the yclm gene coding for aspartate kinase III is regulated by threonine and lysine [25, 48, 80]. To our surprise, the band of full-length transcript turned out to be rather low. What could be the reason for their relatively low amount the lysc mrna? As mentioned, B. subtilis codes for a total of three aspartate kinase iso enzymes. Why there are three enzymes (as also described for E. coli [114]) is not known. The possibility exists that there is no need for a significant increase in the amount of aspartate kinase II, since kinase I or III or both are present in sufficient amounts. It will be interesting to analyse transcription of dapg and yclm under condition of high and low L-lysine concentrations. It will also be interesting to delete dapg or yclm and both genes together, and study the influence of these deletion derivatives on the expression of lysc. Perhaps, deletion of dapg and yclm together will significantly enhance expression of lysc under condition of L-lysine depletion. 113

122 Fig The lysine-responsive riboswitch. (A) Secondary structure [153]; (B) primary structure; (C) primary structure with a potential open reading frame. Analysis of the transcriptional fusion between the promoter region of lysc and lacz indicates that it can be used for controlled expression of recombinant genes as already mentioned. Since removal of L-lysine by centrifugation is only feasible with small volumes, expression in large volumes requires a different and less time-consumable method. One possibility is to start 114

123 growth in an appropriate concentration of L-lysine, which, after consumption, leads to autoinduction of the recombinant gene. We have tested this hypothesis and obtained only a 3-fold induction after 4 h of growth starting with 5 μg/ml L-lysine (Fig. 3.21B). An alternative method would involve regulated expression of an artificial aptamer able to quickly sequester the L-lysine upon induction. There are already several examples where engineered riboswitches have been used to obtain regulated gene expression (reviewed in [6]) including an artificial system for B. subtilis, where expression of recombinant genes can be induced by the addition of theophylline [6], but none of them has used a natural riboswitch so far. 4.2 A promoter-probe plasmid for screening promoters in B. subtilis The promoter-probe vector pht06 based on a multi-copy theta-replication plasmid [156, 157] can replace the single-copy pbgab [99] for screening of strong promoters, which allows heat stable β-galactosidase gene expresson in the cytoplasm. Plasmid pht06 can be used as a promoter-probe vector by introducing DNA fragments at BamHI, NheI, SacI, KpnI, SpeI, and SacII or as controllable promoter-test vector as done in this work. In addition, one can use the feature of laco to control expression in E. coli during cloning, but not in B. subtilis by replacing laci by the promoter at SacII, NheI and SacI. As an example, a library of synthetic promoters was introduced, which had various strengths leading to different expression levels. In this work, promoters P88 and P89 were chosen to demonstrate the utility of plasmid pht06 for cloning and screening strong promoters successfully. Two obstacles were overcome during this work: (i) cloning and (ii) conditions to compare the strength of promoters by applying a linear range of IPTG concentrations. First, while it is difficult to use the available promoter-probe vectors pndh05, pt05z (this work), placz [176] or pbgab [99] for cloning of strong promoters, which causes lethal overproduction of recombinant proteins in E. coli, this problem can be avoided by using the new promoter-test plasmid pht06. Second, there is little information about the correlation of IPTG concentrations and β-galactosidase activity. Actually, we faced this problem when all the promoters from the libraries were checked at 1 mm and then at 0.1 mm IPTG (data not shown). It turned out that almost all strong promoters conferred nearly the same activity. For example, the activity of P88, P89 and P grac were nearly the same at 1 mm of IPTG after 2 or 4 h of induction and at 0.01 mm P88 and P89 were about 10 times higher thanthat of P grac. This useful promoter-probe vector could largely solve problems to screen strong promoters by observation of the coulor of colonies on X-gal plates. 115

124 The low copy number and structural stable vector pht06 together with the expression vector pht01 [105] would become interesting for anyone who wants to establish a library of promoters or genes. With these available techniques, one could generate mutant libraries using the error-prone PCR method [27] with these new plasmids to identify and study elements of promoters in B. subtilis or gene functions. 4.3 Improve the productivity of B. subtilis B. subtilis produces and secretes large amount of various proteins into the culture medium [120]. For the last two decades numerous attempts have been made to use this bacterium as an efficient host for the expression of recombinant proteins [147]. To improve the productivity of B. subtilis, we aimed to identify efficient regulatory elements enhancing transcription from expression plasmids. Published data [105, 117] and those presented in Fig prove that the P grac promoter is indeed a strong promoter, which consists not only of the strong P groe promoter, but also of the lac operater with a dual function: control of transcription and acting as a stabilizing element. Further efforts have been focused on the development of elements for increasing the promoter strength and for identification of mrna-stabilizing elements in overproduction of proteins Strong promoters in B. subtilis A compilation and analysis of B. subtilis σ A -dependent promoter sequences [57] indicates that promoter recognition depends on the following consensus sequences: (i) the transcription start site, which should be a purine, (ii) the -10 region TATAAT, (iii) a six nucleotide spacer between the +1 and the -10 region, (iv) a TnTG sequence from -17 to -14, (v) the -35 region TTGACA, (vi) a 17 nucleotide spacer between the -10 and the -35 region and (vii) an upstream region containing An and Tn from -70 to -36 called UP element. Based on this information, I focused on sequences at transcriptional start site, the -10 region, the -15 region, the -35 region and the UP element upstream of the groe promoter aiming to construct stronger promoters. To analyse the elements for a strong promoter, the newly constructed promoter-probe plasmid pht06 was used. A total of 85 different synthetic and groe-modified σ A -dependent promoters (Fig. 3.31) were cloned into pht06 and analysed by measuring the BgaB activities, the amount and stability of the bgab transcripts and the amount of BgaB produced. The promoter recognition by the bacterial RNAP holoenzyme involves interactions not only between core promoter elements and the sigma subunit, but also between a DNA element 116

125 upstream of the core promoter (UP element) and the α subunit. DNA binding by the α subunit can increase transcription dramatically [47]. UP elements are AT-rich sequences located upstream of the -35 element of some promoters [20, 47]. They act as binding sites for the αctd of the α subunit of the RNAP [13, 52]. The promoter P groe precedes the groesl operon coding for the GroEL chaperonin [137] and is a strong promoter because of a potential UP element (Tab. 3.1), and its activity is comparable with the mutant C2mu UP element [96]. The UP element of groe was modified based on the consensus sequence. The groe UP element sequence from P64 (-55-AAAATTTTTTAAAAAATCTC-36) served as an active UP element also for other σ A -dependent promoters, and resulted in a 2-fold increase in the BgaB activity and a 5-fold higher mrna expression as compared to P grac (P01) [105, 117]. This agrees well with the alignment of 236 B. subtilis σ A -dependent promoters [57] demonstrating that the -54-nAnnnnnTnnnAAAAnnnTn-36 sequence represents an UP element of σ A -dependent B. subtilis promoters. As mentioned, UP elements increase the affinity between the RNA polymerase and the downstream core promoter. But, as shown here, when the core promoter provides already a high affinity, it cannot be further increased by addition of an UP element. It follows that an UP element increases the strength of only those promoters with a low affinity for the RNA polymerase. To ensure a high transcription rate, nature has evolved two possibilities: a strong core promoter or a weak core promoter with an UP element. Fig Secondary structure of the laco stem-loop of S01 (in P grac ), S104 (in P104) and S205. When the UGU at transcriptional start site region was changed to AGG (in P104, Tab. 3.3), the BgaB expression level was more than doubled. It can be concluded that the new potential transcriptional start site confers increased stability to the transcript. When the U was replaced by an A residue (P205), this increases the predicted free energy from to -8.7 (Fig. 4.3) 117

126 and reduced the mrna stability. All these data point to the possibility to use the laco as a 5 stabilizing element. One set of mutations, located between positions -35 and -10, was created and tested. It turned out that mutation of the extended -10 region (named -15 region in this thesis) was very important (Tab. 3.4, P73). The term extended 10 promoters refers to the TG motif, 1 base upstream of the 10 region, which plays a vital role in several E.coli promoters [76]. In B. subtilis, it is termed the -16 region with the sequence 5 -RTRTG-3 [100]. Then, a more comprehensive analysis of 142 promoters, all with experimentally determined transcription start sites, confirm that the 16 region (TRTG) is conserved [57]. Based on this analysis, the -15 region TTCT in the P WT groe promoter was replaced by TATG, in P73, and the BgaB activity increased up to 94-fold. This result strengthens those published by Voskuil, indicating that the -16 region TRTG motif (R = purine) appears to be a basic element present in many gram-positive bacterial promoters [162]. We and others noticed that the -16 region TRTG motif was essential in some promoters [23, 89, 162, 171], especially in weak promoters [59] or promoters that lack an UP element [18, 19], or promoters that lack an identifiable 35 region [9, 22, 76, 162]. Promoters with a combination of UP elements and consensus core promoter elements (-15 ATG and -10 consensus; -15 ATG and -35 consensus; or -15 ATG and -10 and -35 consensus) resulted in the production of BgaB protein reaching levels of up to about 30% of the total cellular protein, especially with those promoters containing the -15 ATG triplet Role of messenger RNA stabilizing elements in overproduction of proteins Gene expression levels are mainly determined by the efficiency of transcription, mrna stability and the frequency of mrna translation. Transcription (in 4.3.1) and translation has been the subject of intense optimization in recombinant expression systems. Stability of the mrna transcript is however rarely addressed. But gene expression is controlled by the decay of mrna [123, 124]. Messenger RNA stability has been shown to play an important role in the regulation of gene expression, and much needs to be learned about the elements that function as mrna stability determinants. Four different elements with stabilizing potentialities will be discussed separately, namely: (i) a 3 mrna terminal stem-loop; (ii) the 5 laco stem-loop; (iii) a strong RBS; (iv) the spacer length between the 5 laco stem-loop and the RBS. 118

127 3 mrna terminal stem-loop structures We found that the terminal stem-loop at the 3 end of the mrna did not correlate with mrna stability (Tab. 3.6, Fig. 3.38). This is in contrast to findings of Wong et al. (1986), who concluded, based on their data, that the 3 end of the cry gene of B. thuringiensis (Fig. 4.4) conferred increased stability on other mrnas in both E. coli and B. subtilis [169]. In another publication, addition of the cry transcription terminator at the 3' end of the lacz gene did not confer any increase in its stability and produced only marginal effects on the final β-galactosidase activity [70]. This finding indicates that such a 3 stabilizing element does not act by itself, but is influenced by the sequence of the transcript in a way, which is not yet understood at the molecular level. One possibility would be the formation of alternative secondary structures. A spacer region, inserted between the 3 end of the mrna and the cry element could alleviate this problem. Such a spacer region, which by itself will form a stemloop structures could protect the cry hairpin structure or any other additional secondary structure. The average half-life of the bgab transcripts was 30 min in this study with modified versions of the trpa terminator or the natural terminators of the skfa, htpg and dnak genes. These results indicate that the 3 -mrna terminal stem-loop structures did not further increase the half-life of mrnas as compared to P01 with the trpa WT terminator (ΔG = kcal/mol) with a half-life about 30 min. Wong [169] employed the penp gene with a half-life of min to study the influence of the cry terminator in both E. coli and B. subtilis. The data in Tab. 3.6 indicate that the E. coli trpa terminator had only a slight influence on the BgaB activity. These results are consistent with the results of Wong [169], who had tested the terminators from two additional bacterial genes that encode stable mrnas, the lpp gene of E. coli and the ery gene from the S. aureus. The mrna half-lifes for these transcripts were reported to be 11.5 and >22 min, respectively. However, no enhancement on the expression of the penp gene was obtained; the penp-lpp and penp-ery fusions expressed penp at the same level as did the native penp gene in both E. coli and B. subtilis. These findings indicate that the half-life of a transcript and its translation efficiency do not necessarily correlate. This is surprising and further indicates that there might be an additional mechanism(s) influencing the amount of protein to be discovered. 119

128 Fig Secondary structure of the potential 3'-end of the cry mrna [169]. We noted that after removal of some T residues (U residues in Tab. 3.6), the BgaB activity increased. This means that leaky transcription termination of the upstream located gene orf-1 (called repa) encoding the plasmid replication initiator [156] would increase BgaB expression. Increasing the amount of RepA in turn should lead to an increase in the copy number of the plasmids with a concomitant enhanced production of the BgaB protein due to the gene dosin effect. However, there is a transcriptional terminator just before orf-1; only a small amount of the transcript could reach to orf-1 due to overriding the transcriptional terminator. One might exploit this feature to conditionally increase the copy number of plasmid for protein overproduction purpose. On the other hand, the exoribonuclease PNPase activity is demonstrated in B. subtilis by Deutscher and Reuven [34] and shown to be the predominant degradative activity in cell extracts with poly(a) or poly(u) RNA as a substrate. The 5 laco stem-loop structure The results shown in Tab. 3.7 and Fig point to the importance of a stem-loop as exemplified by laco. Therefore, the effect of the predicted laco stem-loop on mrna stability was analysed. When constructs with decreased free energy of laco by changing nucleotides inside of laco (Fig.4.5, S211) were tested, the BgaB activity increased more than 2-fold as compared to the P grac promoter (Tab. 3.7, Fig. 3.39, S206 and S211). This is consistent with the idea that the major determinant for mrna stability is a blocked 5 -end [146]. The laco stem-loop increased the BgaB activity >2-fold, but there was no clear effection mrna stability. Therefore, it would be of interest to determine whether a strong RBS and a spacer between the 5 stem-loop and the RBS caused an effect in the half-life of the mrna. 120

129 Fig Secondary structure of the 5 stem-loop from S01 (in P grac ) and S211 Influence of a strong RBS on mrna stability We found that the strong RBS of gsib correlated with mrna stability and expression of the reporter gene (Tab. 3.8 and Fig and Fig. 3.42), when it was fused to P lepa. This is in agreement with the findings of B. thuringiensis cryiiia mrna [1] and gsib mrna [74] that mrna stability can depend on a strong RBS, which is located four (cryiiia) or nine (gsib) nucleotides away from the 5 -end. These results suggest that a strong RBS enhance mrna stability when positioned close the 5 end, most probably by impairing binding of an endonuclease. The spacer length between 5 the stem-loop and the RBS The experiments shown in Fig and 3.40 demonstrated that even a highly stable 5 -terminal structure could not confer stability without an adequate distance between the 5 end of the structure and the RBS. The spacer of a length of 13 to 29 bp not only influenced the mrna stability, but also the BgaB protein expression level (Tab. 3.9 and Fig. 3.40). We and other propose that the binding of ribosomes at an RBS that is located too close to the 5 -terminal structure will result in perturbation of the structure. Based on crystallography studies [175], 15 nts upstream of the initiation codon are protected by the ribosome bound in a ternary complex. In the case of this study, a spacer length of 19 nucleotides in the case of S213 can be considered as optimal for production of BgaB, in that the BgaB activity could be increased up to 9-fold as compared to the P grac promoter with a half-life of the mrna of more than 60 min (Fig. 3.40). These data suggests that this spacer is short enough to allow ribosome binding without being inhibited by the formation of the protective 5 -terminal 121

130 structure. Once the spacer is increased to 29 nts (S110, Fig. 3.40), ribosome binding no longer affected secondary structure formation, resulting in a very stable mrna. It can be concluded that the laco stem-loop, a Controllable Stabilizing Element (CoSE) and elements of strong σ A -dependent P groe promoter can be used for overproduction of recombinant proteins in B. subtilis. To evaluate its potential, CoSE was combined with other promoters, and all constructs could increase the half-life of the mrna and increase the BgaB activity and the yield of BgaB, PBP4* and HtpG proteins. Promoter P223 consists of CoSE S212 and the strong promoter P68 able to overproduce BgaB up to 42% of total cellular protein (Fig. 3.42). While BgaB protein from strong P grac 9.2% [105, 117]; from P WT groe - CoSE, S212, 26.7% (Fig. 3.42); from P68 and P100 (see in Fig 3.35), BgaB up to 30% of total cellular protein. The other recombinant proteins, PBP4*, P100 could overproduce this protein up to 38% and P250 up to 28% in total cellular protein (16% in P grac ); HtpG, P250 could overproduce this protein up to 26% in total cellular protein (12% in P grac ). On the other hand, CoSE could be extended to other bacteria. See in Fig. 4.6, the positive effect on BgaB expression in E. coli, the result shown that S01 (pht01) is weaker than S212 and S221 is weaker than S224, which is similar the result in B. subtilis. Fig Using CoSE in E. coli. Cell with different plasmids/stabilizing element were stretched on X-gal LB plate with 0.01 mm IPTG. In summary, all of strong promoters and CoSE explored in this study are extremely useful for improvement expression system in both B. subtilis and E. coli. 122

131 4.4 Outlook B. subtilis is used to produce and to secrete large amounts of various proteins into the culture medium since the last two decades [120]. Then, numerous attempts have been made to convert this bacterium into an efficient host for the expression of recombinant genes [147]. At present, about 60% of the commercially available enzymes are produced by Bacillus species; and a large body of information concerning transcription, translation, protein folding and secretion mechanisms, genetic manipulation, and large-scale fermentation has been acquired as mentioned in the Introduction of this thesis and in [56, 97, 163, 170]. But more knowledge on cellular functions and development of better production systems is still needed [163]. Based on the results in this thesis, a key question is how the glycine-inducible and the lysine-autoinducible expression systems can be improved for industrial production and how strong promoters with CoSE as a useful tool can be applied for the downstream processing. (i) One may improve the glycine-inducible system by fusing the CoSE at downstream of the riboswitch and to combine it with a strong promoter. This system can be used for overproduction of recombinant proteins using the inexpensive inducer glycine. (ii) As discussed, there is the possibility to improve the expression level of the lysineresponsive riboswitch-regulated lysc gene by deleting the two other genes coding for aspatokinase iso enzymes. These experiments are in progress. (iii) One possibility to reduce basal level of expression in E. coli in to use shuttle vectors with a low copy number in E. coli and a high copy number in B. subtilis. (iv) Based on the strong promoter P68, P100, S212, S223 or S250 expression vectors with purification tags and secretion vectors can be developed. 123

132 5 References [1] Agaisse, H. and Lereclus, D. (1996) STAB-SD: a Shine-Dalgarno sequence in the 5' untranslated region is a determinant of mrna stability. Mol. Microbiol. 20, [2] Airaksinen, U., Penttila, T., Wahlstrom, E., Vuola, J.M., Puolakkainen, M. and Sarvas, M. (2003) Production of Chlamydia pneumoniae proteins in Bacillus subtilis and their use in characterizing immune responses in the experimental infection model. Clin. Diagn. Lab Immunol. 10, [3] Alifano, P., Bruni, C.B. and Carlomagno, M.S. (1994) Control of mrna processing and decay in prokaryotes. Genetica. 94, [4] Anderson, R.P. and Roth, J.R. (1977) Tandem genetic duplications in phage and bacteria. Annu. Rev. Microbiol. 31, [5] Barrick, J.E., Corbino, K.A., Winkler, W.C., Nahvi, A., Mandal, M., Collins, J., Lee, M., Roth, A., Sudarsan, N., Jona, I., Wickiser, J.K. and Breaker, R.R. (2004) New RNA motifs suggest an expanded scope for riboswitches in bacterial genetic control. Proc. Natl. Acad. Sci. USA 101, [6] Bauer, G. and Suess, B. (2006) Engineered riboswitches as novel tools in molecular biology. J. Biotechnol. 124, [7] Bechhofer, D.H. and Dubnau, D. (1987) Induced mrna stability in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 84, [8] Bechhofer, D.H. and Zen, K.H. (1989) Mechanism of erythromycin-induced ermc mrna stability in Bacillus subtilis. J. Bacteriol. 171, [9] Belyaeva, T., Griffiths, L., Minchin, S., Cole, J. and Busby, S. (1993) The Escherichia coli cysg promoter belongs to the 'extended -10' class of bacterial promoters. Biochem. J. 296 (Pt 3), [10] Bhavsar, A.P., Zhao, X. and Brown, E.D. (2001) Development and characterization of a xylose-dependent system for expression of cloned genes in Bacillus subtilis: conditional complementation of a teichoic acid mutant. Appl. Environ. Microbiol. 67, [11] Bill Campbell. (1997) Purification Tricks for Recombinant Proteins. [email protected]. Webpage. [12] BioCoach Activity. (2007) Transcription and Translation in Cells. Webpage. [13] Birnboim, H.C. and Doly, J. (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7,

133 [14] Bongers, R.S., Veening, J.W., Van, W.M., Kuipers, O.P. and Kleerebezem, M. (2005) Development and characterization of a subtilin-regulated expression system in Bacillus subtilis: strict control of gene expression by addition of subtilin. Appl. Environ. Microbiol. 71, [15] Bouvet, P. and Belasco, J.G. (1992) Control of RNase E-mediated RNA degradation by 5'-terminal base pairing in E. coli. Nature 360, [16] Bricker, A.L. and Belasco, J.G. (1999) Importance of a 5' stem-loop for longevity of papa mrna in Escherichia coli. J. Bacteriol. 181, [17] Bron, S., Bolhuis, A., Tjalsma, H., Holsappel, S., Venema, G. and van Dijl, J.M. (1998) Protein secretion and possible roles for multiple signal peptidases for precursor processing in bacilli. J. Biotechnol. 64, [18] Burns, H. and Minchin, S. (1994) Thermal energy requirement for strand separation during transcription initiation: the effect of supercoiling and extended protein DNA contacts. Nucleic Acids Res. 22, [19] Burns, H.D., Belyaeva, T.A., Busby, S.J. and Minchin, S.D. (1996) Temperaturedependence of open-complex formation at two Escherichia coli promoters with extended -10 sequences. Biochem. J. 317 (Pt 1), [20] Busby, S. and Ebright, R.H. (1994) Promoter structure, promoter recognition, and transcription activation in prokaryotes. Cell 79, [21] Caramori, T. and Galizzi, A. (1998) The UP element of the promoter for the flagellin gene, hag, stimulates transcription from both SigD- and SigA-dependent promoters in Bacillus subtilis. Mol. Gen. Genet. 258, [22] Chan, B., Spassky, A. and Busby, S. (1990) The organization of open complexes between Escherichia coli RNA polymerase and DNA fragments carrying promoters either with or without consensus -35 region sequences. Biochem. J. 270, [23] Chen, G., Kumar, A., Wyman, T.H. and Moran, C.P., Jr. (2006) Spo0A-dependent activation of an extended -10 region promoter in Bacillus subtilis. J. Bacteriol. 188, [24] Chen, N.Y., Hu, F.M. and Paulus, H. (1987) Nucleotide sequence of the overlapping genes for the subunits of Bacillus subtilis aspartokinase II and their control regions. Biol. Chem. 262: [25] Chen, N.Y., Jiang, S.Q., Klein, D.A. and Paulus, H. (1993) Organization and nucleotide sequence of the Bacillus subtilis diaminopimelate operon, a cluster of genes encoding the first three enzymes of diaminopimelate synthesis and dipicolinate synthase. J. Biol. Chem. 268, [26] Chowdhury, S., Ragaz, C., Kreuger, E. and Narberhaus, F. (2003) Temperaturecontrolled structural alterations of an RNA thermometer. J. Biol. Chem. 278,

134 [27] Cirino, P.C., Mayer, K.M. and Umeno, D. (2003) Generating mutant libraries using error-prone PCR. In: Directed evolution library creation: methods and protocols (Arnold, F.H. and Georgiou, G., Eds.), Vol. 231, pp Humana press. [28] Condon, C. (2003) RNA processing and degradation in Bacillus subtilis. Microbiol. Mol. Biol. Rev. 67, [29] Conrad, B., Savchenko, R.S., Breves, R. and Hofemeister, J. (1996) A T7 promoterspecific, inducible protein expression system for Bacillus subtilis. Mol. Gen. Genet. 250, [30] Cornet, P., Millet, J., Beguin, P. and Aubert, J.P. (1983) Characterization of two cel (cellulose degradation) genes of Clostridium thermocellum coding for endoglucanases. Bio/Technology 1, [31] Cutting, S., Anderson, M., Lysenko, E., Page, A., Tomoyasu, T., Tatematsu, K., Tatsuta, T., Kroos, L. and Ogura, T. (1997) SpoVM, a small protein essential to development in Bacillus subtilis, interacts with the ATP-dependent protease FtsH. J. Bacteriol. 179, [32] Daguer, J.P., Chambert, R. and Petit-Glatron, M.F. (2005) Increasing the stability of sacb transcript improves levansucrase production in Bacillus subtilis. Lett. Appl. Microbiol. 41, [33] Del, S.G. and Espinosa, M. (2000) Plasmid copy number control: an ever-growing story. Mol. Microbiol. 37, [34] Deutscher, M.P. and Reuven, N.B. (1991) Enzymatic basis for hydrolytic versus phosphorolytic mrna degradation in Escherichia coli and Bacillus subtilis. Proc. Natl. Acad. Sci. USA 88, [35] Donovan, W.P. and Kushner, S.R. (1986) Polynucleotide phosphorylase and ribonuclease II are required for cell viability and mrna turnover in Escherichia coli K-12. Proc. Natl. Acad. Sci. USA 83, [36] Dower, W.J., Miller, J.F. and Ragsdale, C.W. (1988) High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16, [37] Ellington, A.D. and Szostak, J.W. (1990) In vitro selection of RNA molecules that bind specific ligands. Nature 346, [38] Emory, S.A., Bouvet, P. and Belasco, J.G. (1992) A 5'-terminal stem-loop structure can stabilize mrna in Escherichia coli. Genes Dev. 6, [39] Evan, G.I., Lewis, G.K., Ramsay, G. and Bishop, J.M. (1985) Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product. Mol. Cell Biol. 5, [40] Fukushima, T., Yamamoto, H., Atrih, A., Foster, S.J. and Sekiguchi, J. (2002) A polysaccharide deacetylase gene (pdaa) is required for germination and for production of muramic delta-lactam residues in the spore cortex of Bacillus subtilis. J. Bacteriol. 184,

135 [41] Gail, M.E. (2004) Riboswitch Regulation of Gene Expression. Yale University. Webpage. [42] Geissendorfer, M. and Hillen, W. (1990) Regulated expression of heterologous genes in Bacillus subtilis using the Tn10 encoded Tet regulatory elements. Appl. Microbiol. Biotechnol. 33, [43] Ghosh, S. and Deutscher, M.P. (1999) Oligoribonuclease is an essential component of the mrna decay pathway. Proc. Natl. Acad. Sci. USA 96, [44] Glatz, E., Nilsson, R.P., Rutberg, L. and Rutberg, B. (1996) A dual role for the Bacillus subtilis glpd leader and the GlpP protein in the regulated expression of glpd: antitermination and control of mrna stability. Mol. Microbiol. 19, [45] Glatz, E., Persson, M. and Rutberg, B. (1998) Antiterminator protein GlpP of Bacillus subtilis binds to glpd leader mrna. Microbiology 144 (Pt 2), [46] Gottesman, S. (2004) The small RNA regulators of Escherichia coli: roles and mechanisms. Annu. Rev. Microbiol. 58, [47] Gourse, R.L., Ross, W. and Gaal, T. (2000) UPs and downs in bacterial transcription initiation: the role of the alpha subunit of RNA polymerase in promoter recognition. Mol. Microbiol. 37, [48] Graves, L.M. and Switzer, R.L. (1990) Aspartokinase III, a new isozyme in Bacillus subtilis 168. J. Bacteriol. 172, [49] Graves, M.C. and Rabinowitz, J.C. (1986) In vivo and in vitro transcription of the Clostridium pasteurianum ferredoxin gene. Evidence for "extended" promoter elements in gram-positive organisms. J. Biol. Chem. 261, [50] Grundy, F.J. and Henkin, T.M. (2006) From ribosome to riboswitch: control of gene expression in bacteria by RNA structural rearrangements. Crit Rev. Biochem. Mol. Biol. 41, [51] Grundy, F.J., Lehman, S.C. and Henkin, T.M. (2003) The L box regulon: lysine sensing by leader RNAs of bacterial lysine biosynthesis genes. Proc. Natl. Acad. Sci. USA 100, [52] Guerout-Fleury, A.M., Frandsen, N. and Stragier, P. (1996) Plasmids for ectopic integration in Bacillus subtilis. Gene 180, [53] Hambraeus, G., Karhumaa, K. and Rutberg, B. (2002) A 5' stem-loop and ribosome binding but not translation are important for the stability of Bacillus subtilis apre leader mrna. Microbiology 148, [54] Hambraeus, G., Persson, M. and Rutberg, B. (2000) The apre leader is a determinant of extreme mrna stability in Bacillus subtilis. Microbiology 146 (Pt 12), [55] Härtl, B., Wehrl, W., Wiegert, T., Homuth, G. and Schumann, W. (2001) Development of a new integration site within the Bacillus subtilis chromosome and construction of compatible expression cassettes. J. Bacteriol. 183,

136 [56] Harwood, C.R. (1992) Bacillus subtilis and its relatives: molecular biological and industrial workhorses. Trends Biotechnol. 10, [57] Helmann, J.D. (1995) Compilation and analysis of Bacillus subtilis sigma A- dependent promoter sequences: evidence for extended contact between RNA polymerase and upstream promoter DNA. Nucleic Acids Res. 23, [58] Helmann, J.D. and Moran Jr, C.P. (2002) RNA polymerase and sigma factors. In: Bacillus subtilis and its closest relatives (Sonenshein, A.L., Hoch, J.A. and Losick, R., Eds.), pp ASM Press, Washington, D.C. [59] Henkin, T.M. and Sonenshein, A.L. (1987) Mutations of the Escherichia coli lacuv5 promoter resulting in increased expression in Bacillus subtilis. Mol. Gen. Genet. 209, [60] Herman, C., Thevenet, D., D'Ari, R. and Bouloc, P. (1997) The HflB protease of Escherichia coli degrades its inhibitor lambda ciii. J. Bacteriol. 179, [61] Hirata, H., Fukazawa, T., Negoro, S. and Okada, H. (1986) Structure of a betagalactosidase gene of Bacillus stearothermophilus. J. Bacteriol. 166, [62] Hirata, H., Negoro, S. and Okada, H. (1985) High Production of Thermostable betagalactosidase of Bacillus stearothermophilus in Bacillus subtilis. Appl. Environ. Microbiol. 49, [63] Ho, K.M. and Lim, B.L. (2003) Co-expression of a prophage system and a plasmid system in Bacillus subtilis. Protein Expr. Purif. 32, [64] Hochuli, E. (1990) Purification of recombinant proteins with metal chelate adsorbent. Genet. Eng. 12, [65] Homuth, G., Masuda, S., Mogk, A., Kobayashi, Y. and Schumann, W. (1997) The dnak operon of Bacillus subtilis is heptacistronic. J. Bacteriol. 179, [66] Huang, H., Ridgway, D., Gu, T. and Moo-Young, M. (2004) Enhanced amylase production by Bacillus subtilis using a dual exponential feeding strategy. Bioprocess. Biosyst. Eng 27, [67] IBA. (2005) Strep-Tactin Spin Column Purification Protocol. Handbook. [68] Inoue, H., Nojima, H. and Okayama, H. (1990) High efficiency transformation of Escherichia coli with plasmids. Gene 96, [69] Ish-Horowicz, D. and Burke, J.F. (1981) Rapid and efficient cosmid cloning. Nucleic Acids Res. 9, [70] Jan, J., Valle, F., Bolivar, F. and Merino, E. (2001) Construction of protein overproducer strains in Bacillus subtilis by an integrative approach. Appl. Microbiol. Biotechnol. 55, [71] Jan, J., Valle, F., Bolivar, F. and Merino, E. (2000) Characterization of the 5' subtilisin (apre) regulatory region from Bacillus subtilis. FEMS Microbiol. Lett. 183,

137 [72] Johansson, J., Mandin, P., Renzoni, A., Chiaruttini, C., Springer, M. and Cossart, P. (2002) An RNA thermosensor controls expression of virulence genes in Listeria monocytogenes. Cell 110, [73] Junttila, M.R., Saarinen, S., Schmidt, T., Kast, J. and Westermarck, J. (2005) Singlestep Strep-tag purification for the isolation and identification of protein complexes from mammalian cells. Proteomics. 5, [74] Jürgen, B., Schweder, T. and Hecker, M. (1998) The stability of mrna from the gsib gene of Bacillus subtilis is dependent on the presence of a strong ribosome binding site. Mol. Gen. Genet. 258, [75] Kaltwasser, M., Wiegert, T. and Schumann, W. (2002) Construction and application of epitope- and green fluorescent protein-tagging integration vectors for Bacillus subtilis. Appl. Environ. Microbiol. 68, [76] Keilty, S. and Rosenberg, M. (1987) Constitutive function of a positively regulated promoter reveals new sequences essential for activity. J. Biol. Chem. 262, [77] Kierzek, A.M., Zaim, J. and Zielenkiewicz, P. (2001) The effect of transcription and translation initiation frequencies on the stochastic fluctuations in prokaryotic gene expression. J. Biol. Chem. 276, [78] Kikuchi, G. (1973) The glycine cleavage system: composition, reaction mechanism, and physiological significance. Mol Cell Biochem. 1, [79] Kim, L., Mogk, A. and Schumann, W. (1996) A xylose-inducible Bacillus subtilis integration vector and its application. Gene 181, [80] Kobashi, N., Nishiyama, M. and Yamane, H. (2001) Characterization of aspartate kinase III of Bacillus subtilis. Biosci. Biotechnol. Biochem. 65, [81] Kruger N.J. (2002) The Bradford method for protein quantitation. In: Protein protocols handbook (Walker, J.M., Ed.), pp Humana Press. [82] Krummel, B. and Chamberlin, M.J. (1989) RNA chain initiation by Escherichia coli RNA polymerase. Structural transitions of the enzyme in early ternary complexes. Biochemistry 28, [83] Kubodera, T., Watanabe, M., Yoshiuchi, K., Yamashita, N., Nishimura, A., Nakai, S., Gomi, K. and Hanamoto, H. (2003) Thiamine-regulated gene expression of Aspergillus oryzae thia requires splicing of the intron containing a riboswitch-like domain in the 5'-UTR. FEBS Lett. 555, [84] Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, [85] Le Grice, S.F. (1990) Regulated promoter for high-level expression of heterologous genes in Bacillus subtilis. Methods Enzymol. 185, [86] Lee, J.F., Hesselberth, J.R., Meyers, L.A. and Ellington, A.D. (2004) Aptamer database. Nucleic Acids Res. 32, D

138 [87] Lee, J.K., Edwards, C.W. and Hulett, F.M. (1991) Bacillus licheniformis APase I gene promoter: a strong well-regulated promoter in B. subtilis. J. Gen. Microbiol. 137, [88] Liu, H.B., Chui, K.S., Chan, C.L., Tsang, C.W. and Leung, Y.C. (2004) An efficient heat-inducible Bacillus subtilis bacteriophage 105 expression and secretion system for the production of the Streptomyces clavuligerus beta-lactamase inhibitory protein (BLIP). J. Biotechnol. 108, [89] Liu, M., Tolstorukov, M., Zhurkin, V., Garges, S. and Adhya, S. (2004) A mutant spacer sequence between -35 and -10 elements makes the Plac promoter hyperactive and camp receptor protein-independent. Proc. Natl. Acad. Sci. USA 101, [90] Mackie, G.A. (2000) Stabilization of circular rpst mrna demonstrates the 5'-end dependence of RNase E action in vivo. J. Biol. Chem. 275, [91] Mandal, M., Boese, B., Barrick, J.E., Winkler, W.C. and Breaker, R.R. (2003) Riboswitches control fundamental biochemical pathways in Bacillus subtilis and other bacteria. Cell 113, [92] Mandal, M. and Breaker, R.R. (2004) Adenine riboswitches and gene activation by disruption of a transcription terminator. Nat. Struct. Mol. Biol. 11, [93] Mandal, M., Lee, M., Barrick, J.E., Weinberg, Z., Emilsson, G.M., Ruzzo, W.L. and Breaker, R.R. (2004) A glycine-dependent riboswitch that uses cooperative binding to control gene expression. Science 306, [94] Maul, B., Volker, U., Riethdorf, S., Engelmann, S. and Hecker, M. (1995) sigma B- dependent regulation of gsib in response to multiple stimuli in Bacillus subtilis. Mol. Gen. Genet. 248, [95] McDaniel, B.A., Grundy, F.J., Artsimovitch, I. and Henkin, T.M. (2003) Transcription termination control of the S box system: direct measurement of S-adenosylmethionine by the leader RNA. Proc. Natl. Acad. Sci. USA 100, [96] Meijer, W.J. and Salas, M. (2004) Relevance of UP elements for three strong Bacillus subtilis phage phi29 promoters. Nucleic Acids Res. 32, [97] Meima, R., van Dijl, J.M. and Bron, S. (2004) Expression systems in Bacillus. In: Protein expression technologies. pp Horizon Bioscience, Norfold, U.K. [98] Ming-Ming, Y., Wei-Wei, Z., Xi-Feng, Z. and Pei-Lin, C. (2006) Construction and characterization of a novel maltose inducible expression vector in Bacillus subtilis. Biotechnol. Lett. 28, [99] Mogk, A., Hayward, R. and Schumann, W. (1996) Integrative vectors for constructing single-copy transcriptional fusions between Bacillus subtilis promoters and various reporter genes encoding heat-stable enzymes. Gene 182, [100] Moran, C.P., Jr., Johnson, W.C. and Losick, R. (1982) Close contacts between sigma 37-RNA polymerase and a Bacillus subtilis chromosomal promoter. J. Mol. Biol. 162,

139 [101] Moran, C.P., Jr., Lang, N., LeGrice, S.F., Lee, G., Stephens, M., Sonenshein, A.L., Pero, J. and Losick, R. (1982) Nucleotide sequences that signal the initiation of transcription and translation in Bacillus subtilis. Mol. Gen. Genet. 186, [102] Morita, M.T., Tanaka, Y., Kodama, T.S., Kyogoku, Y., Yanagi, H. and Yura, T. (1999) Translational induction of heat shock transcription factor sigma-32: evidence for a built-in RNA thermosensor. Genes Dev. 13, [103] Mujacic, M., Cooper, K.W. and Baneyx, F. (1999) Cold-inducible cloning vectors for low-temperature protein expression in Escherichia coli: application to the production of a toxic and proteolytically sensitive fusion protein. Gene 238, [104] Nguyen, H.D., Nguyen, Q.A., Ferreira, R.C., Ferreira, L.C., Tran, L.T. and Schumann, W. (2005) Construction of plasmid-based expression vectors for Bacillus subtilis exhibiting full structural stability. Plasmid 54, [105] Nguyen, H.D., Phan, T.T.P. and Schumann, W. (2006) Expression Vectors for the Rapid Purification of Recombinant Proteins in Bacillus subtilis. Current Microbiology, in press. [106] Nguyen, H.D. and Schumann, W. (2006) Establishment of an experimental system allowing immobilization of proteins on the surface of Bacillus subtilis cells. J. Biotechnol. 122, [107] Nicholson, W.L. and Chambliss, G.H. (1985) Isolation and characterization of a cisacting mutation conferring catabolite repression resistance to alpha-amylase synthesis in Bacillus subtilis. J. Bacteriol. 161, [108] Nudler, E. and Mironov, A.S. (2004) The riboswitch control of bacterial metabolism. Trends Biochem. Sci. 29, [109] Olmos-Soto, J. and Contreras-Flores, R. (2003) Genetic system constructed to overproduce and secrete proinsulin in Bacillus subtilis. Appl. Microbiol. Biotechnol. 62, [110] Ow, M.C., Perwez, T. and Kushner, S.R. (2003) RNase G of Escherichia coli exhibits only limited functional overlap with its essential homologue, RNase E. Mol. Microbiol. 49, [111] Pallen, M.J., Lam, A.C., Antonio, M. and Dunbar, K. (2001) An embarrassment of sortases - a richness of substrates? Trends Microbiol. 9, [112] Palva, I. (1982) Molecular cloning of alpha-amylase gene from Bacillus amyloliquefaciens and its expression in B. subtilis. Gene 19, [113] Palva, I., Lehtovaara, P., Kaariainen, L., Sibakov, M., Cantell, K., Schein, C.H., Kashiwagi, K. and Weissmann, C. (1983) Secretion of interferon by Bacillus subtilis. Gene 22, [114] Paulus, H. (1993) Biosynthesis of the aspartate family of amino acids. In: Bacillus subtilis and Other Gram-Positive Bacteria (Sonenshein, A.L., Hoch, J.A. and Losick, R., Eds.), pp American Society for Microbiology, Washington, D.C. 131

140 [115] Perego, M. and Brannigan, J.A. (2001) Pentapeptide regulation of aspartyl-phosphate phosphatases. Peptides 22, [116] Phan, T.T. and Schumann, W. (2007) Development of a glycine-inducible expression system for Bacillus subtilis. J. Biotechnol. 128, [117] Phan, T.T.P., Nguyen, H.D. and Schumann, W. (2006) Novel plasmid-based expression vectors for intra- and extracellular production of recombinant proteins in Bacillus subtilis. Protein Expr. Purif. 46, [118] Popham, D.L. and Setlow, P. (1993) Cloning, nucleotide sequence, and regulation of the Bacillus subtilis pbpe operon, which codes for penicillin-binding protein 4* and an apparent amino acid racemase. J. Bacteriol. 175, [119] Price, C.W., Gitt, M.A. and Doi, R.H. (1983) Isolation and physical mapping of the gene encoding the major sigma factor of Bacillus subtilis RNA polymerase. Proc. Natl. Acad. Sci. USA 80, [120] Priest, F.G. (1977) Extracellular enzyme synthesis in the genus Bacillus. Bacteriol. Rev. 41, [121] QIAGEN. (2003) Ni-NTA Spin Kit and Ni-NTA Spin Columns. Handbook. [122] Qing, G., Ma, L.C., Khorchid, A., Swapna, G.V., Mal, T.K., Takayama, M.M., Xia, B., Phadtare, S., Ke, H., Acton, T., Montelione, G.T., Ikura, M. and Inouye, M. (2004) Cold-shock induced high-yield protein production in Escherichia coli. Nat. Biotechnol. 22, [123] Rauhut, R. and Klug, G. (1999) mrna degradation in bacteria. FEMS Microbiol. Rev. 23, [124] Regnier, P. and Arraiano, C.M. (2000) Degradation of mrna in bacteria: emergence of ubiquitous features. Bioessays 22, [125] Ripio, M.T., Dominguez-Bernal, G., Suarez, M., Brehm, K., Berche, P. and Vazquez- Boland, J.A. (1996) Transcriptional activation of virulence genes in wild-type strains of Listeria monocytogenes in response to a change in the extracellular medium composition. Res. Microbiol. 147, [126] Robert E.F.J. (1998) RNA isolation strategies. In: RNA methodologies: A laboratory guide for isolation and characterization, pp Academic Press. [127] Rocha, E.P., Danchin, A. and Viari, A. (1999) Translation in Bacillus subtilis: roles and trends of initiation and termination, insights from a genome analysis. Nucleic Acids Res. 27, [128] Roche company. (2003) DIG Application Manual for Filter Hybridization. Hanbook. [129] Ross, W., Gosink, K.K., Salomon, J., Igarashi, K., Zou, C., Ishihama, A., Severinov, K. and Gourse, R.L. (1993) A third recognition element in bacterial promoters: DNA binding by the alpha subunit of RNA polymerase. Science 262,

141 [130] Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, [131] Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N. (1985) Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230, [132] Saito, H., Shibata, T. and Ando, T. (1979) Mapping of genes determining nonpermissiveness and host-specific restriction to bacteriophages in Bacillus subtilis Marburg. Mol. Gen. Genet. 170, [133] Sambrook, J. and Russell, D.W. (2001) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press. [134] Sandler, P. and Weisblum, B. (1989) Erythromycin-induced ribosome stall in the erma leader: a barricade to 5'-to-3' nucleolytic cleavage of the erma transcript. J. Bacteriol. 171, [135] Sashital, D.G. and Butcher, S.E. (2006) Flipping off the riboswitch: RNA structures that control gene expression. ACS Chem. Biol. 1, [136] Schallmey, M., Singh, A. and Ward, O.P. (2004) Developments in the use of Bacillus species for industrial production. Can. J. Microbiol. 50, [137] Schmidt, A., Schiesswohl, M., Völker, U., Hecker, M. and Schumann, W. (1992) Cloning, sequencing, mapping, and transcriptional analysis of the groesl operon from Bacillus subtilis. J. Bacteriol. 174, [138] Schmidt, T.G., Koepke, J., Frank, R. and Skerra, A. (1996) Molecular interaction between the Strep-tag affinity peptide and its cognate target, streptavidin. J. Mol. Biol. 255, [139] Schmidt, T.G. and Skerra, A. (1993) The random peptide library-assisted engineering of a C-terminal affinity peptide, useful for the detection and purification of a functional Ig Fv fragment. Protein Eng. 6, [140] Schmidt, T.G. and Skerra, A. (1994) One-step affinity purification of bacterially produced proteins by means of the "Strep tag" and immobilized recombinant core streptavidin. J. Chromatogr. A 676, [141] Scholz, O., Thiel, A., Hillen, W. and Niederweis, M. (2000) Quantitative analysis of gene expression with an improved green fluorescent protein. Eur. J Biochem. 267, [142] Schön, U. and Schumann, W. (1994) Construction of His6-tagging vectors allowing single-step purification of GroES and other polypeptides produced in Bacillus subtilis. Gene 147, [143] Schulz, A., Schwab, S., Homuth, G., Versteeg, S. and Schumann, W. (1997) The htpg gene of Bacillus subtilis belongs to class III heat shock genes and is under negative control. J. Bacteriol. 179,

142 [144] Schumann, W. (2007) Production of Recombinant Proteins in Bacillus subtilis. Advances Appl. Microbiol., in press. [145] Sharp, J.S. and Bechhofer, D.H. (2003) Effect of translational signals on mrna decay in Bacillus subtilis. J. Bacteriol. 185, [146] Sharp, J.S. and Bechhofer, D.H. (2005) Effect of 5'-proximal elements on decay of a model mrna in Bacillus subtilis. Mol. Microbiol. 57, [147] Simonen, M. and Palva, I. (1993) Protein secretion in Bacillus species. Microbiol. Rev. 57, [148] Soukup, J.K. and Soukup, G.A. (2004) Riboswitches exert genetic control through metabolite-induced conformational change. Curr. Opin. Struct. Biol. 14, [149] Spizizen, J. (1958) Transformation of biochemically deficient strans of Bacillus subtilis by deoxyribonucleate. Proc. Natl. Acad. Sci. USA 44, [150] Steege, D.A. (2000) Emerging features of mrna decay in bacteria. RNA. 6, [151] Storz, G., Opdyke, J.A. and Zhang, A. (2004) Controlling mrna stability and translation with small, noncoding RNAs. Curr. Opin. Microbiol. 7, [152] Sudarsan, N., Barrick, J.E. and Breaker, R.R. (2003) Metabolite-binding RNA domains are present in the genes of eukaryotes. RNA. 9, [153] Sudarsan, N., Wickiser, J.K., Nakamura, S., Ebert, M.S. and Breaker, R.R. (2003) An mrna structure in bacteria that controls gene expression by binding lysine. Genes Dev. 17, [154] Taira, S., Jalonen, E., Paton, J.C., Sarvas, M. and Runeberg-Nyman, K. (1989) Production of pneumolysin, a pneumococcal toxin, in Bacillus subtilis. Gene 77, [155] Terpe, K. (2003) Overview of tag protein fusions: from molecular and biochemical fundamentals to commercial systems. Appl. Microbiol. Biotechnol. 60, [156] Titok, M.A., Chapuis, J., Selezneva, Y.V., Lagodich, A.V., Prokulevich, V.A., Ehrlich, S.D. and Janniere, L. (2003) Bacillus subtilis soil isolates: plasmid replicon analysis and construction of a new theta-replicating vector. Plasmid 49, [157] Titok, M.A., Suski, C., Dalmais, B., Ehrlich, D.S. and Janniere, L. (2006) The replicative polymerases PolC and DnaE are required for theta replication of the Bacillus subtilis plasmid pbs72. Appl. Environ. Microbiol. In processing. [158] Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA 76, [159] Tucker, B.J. and Breaker, R.R. (2005) Riboswitches as versatile gene control elements. Curr. Opin. Struct. Biol. 15,

143 [160] United States Patent. (2005) No. US 6,911,322 B2, Date Jun. 28, [161] Vagner, V., Dervyn, E. and Ehrlich, S.D. (1998) A vector for systematic gene inactivation in Bacillus subtilis. Microbiology 144, [162] Voskuil, M.I. and Chambliss, G.H. (1998) The -16 region of Bacillus subtilis and other gram-positive bacterial promoters. Nucleic Acids Res. 26, [163] Westers, L., Westers, H. and Quax, W.J. (2004) Bacillus subtilis as cell factory for pharmaceutical proteins: a biotechnological approach to optimize the host organism. Biochim. Biophys. Acta. 1694, [164] Wiegert, T., Homuth, G., Versteeg, S. and Schumann, W. (2001) Alkaline shock induces the Bacillus subtilis sigma(w) regulon. Mol. Microbiol. 41, [165] Wild, J. and Szybalski, W. (2004) Copy-control tightly regulated expression vectors based on pbac/oriv. In: Recombinant gene expression: reviews and protocols (Balbas P. and Lorence A., Eds.), Vol. 267, pp Humana Press, Totowa, NJ. [166] Winkler, W., Nahvi, A. and Breaker, R.R. (2002) Thiamine derivatives bind messenger RNAs directly to regulate bacterial gene expression. Nature 419, [167] Winkler, W.C. and Breaker, R.R. (2003) Genetic control by metabolite-binding riboswitches. Chembiochem. 4, [168] Winkler, W.C., Nahvi, A., Roth, A., Collins, J.A. and Breaker, R.R. (2004) Control of gene expression by a natural metabolite-responsive ribozyme. Nature 428, [169] Wong, H.C. and Chang, S. (1986) Identification of a positive retroregulator that stabilizes mrnas in bacteria. Proc. Natl. Acad. Sci. USA 83, [170] Wong, S.L. (1995) Advances in the use of Bacillus subtilis for the expression and secretion of heterologous proteins. Curr. Opin. Biotechnol. 6, [171] Wosten, M.M., Boeve, M., Koot, M.G., van Nuene, A.C. and van der Zeijst, B.A. (1998) Identification of Campylobacter jejuni promoter sequences. J. Bacteriol. 180, [172] Wu, X.C., Lee, W., Tran, L. and Wong, S.L. (1991) Engineering a Bacillus subtilis expression-secretion system with a strain deficient in six extracellular proteases. J. Bacteriol. 173, [173] Yansura, D.G. and Henner, D.J. (1984) Use of the Escherichia coli lac repressor and operator to control gene expression in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 81, [174] Yeh, E.C. and Steinberg, W. (1978) The effect of gene position, gene dosage and a regulatory mutation on the temporal sequence of enzyme synthesis accompanying outgrowth of Bacillus subtilis spores. Mol. Gen. Genet. 158, [175] Yusupova, G.Z., Yusupov, M.M., Cate, J.H. and Noller, H.F. (2001) The path of messenger RNA through the ribosome. Cell 106,

144 [176] Zellmeier, S., Zuber, U., Schumann, W. and Wiegert, T. (2003) The absence of FtsH metalloprotease activity causes overexpression of the sigmaw-controlled pbpe gene, resulting in filamentous growth of Bacillus subtilis. J. Bacteriol. 185, [177] Zuber, U. and Schumann, W. (1994) CIRCE, a novel heat shock element involved in regulation of heat shock operon dnak of Bacillus subtilis. J. Bacteriol. 176, [178] Zukowski, M.M. and Miller, L. (1986) Hyperproduction of an intracellular heterologous protein in a sacu h mutant of Bacillus subtilis. Gene 46,

145 6 List of abbreviations and symbols Abbreviation Aa A 260 A 280 Amp R amyq APS Denotation amino acid(s) absorption at a wavelength of 260 nm absorption at a wavelength of 280 nm resistant to ampicillin gene coding for protein α-amylase (AmyQ) ammoniumperoxodisulfate α- alpha, indicating antibodies against something (except α- amylase) e.g, α-amyq means antibodies against AmyQ αctd C-terminal domain of the RNA polymerase α subunit B. amyloliquefaciens Bacillus amyloliquefaciens bgab bp coding for the heat stable reporter β-gaclactosidase (BgaB) in G. stearothermophilus base pairs B. subtilis Bacillus subtilis cat cela celb Cm R CMC gene coding for chloraphenicol-acetytransferase gene coding for cellulase A (CelA) from C. thermocellum gene coding for cellulase B (CelB) from C. thermocellum resistant to chloramphenicol carboxy methyl cellulose C. thermocellum Clostridium thermocellum o C DEPC DNA Δ degrees centigrade diethylpyrocarbonate deoxyribonucleic acid deletion E. coli Escherichia coli Et.Br et al. g GFP gly kb kda LB ethidium bromide et alteri gram green fluorescent protein L-glycine kilobase kilo-dalton Luria-Bertani 137

146 lysc h HCl htpg IAA IPTG lacz l MCS min mg ml mm MOPS mrna μg μl Neo nt ON ONPG OD 578 PBS pbpe P grac pmol P spac P gsib P gcv P lysc RBS gene code for the α- and β-subinits of a lysine-responsive aspartokinase II hour(s) hydrocloride acid gene coding for hight temperature protein G (HtpG) isoamylalkohol isopropyl-ß-d-thiogalactoside gene coding for the reporter β-gaclactosidase (LacZ) in E. coli liter Multiple cloning sites minute(s) milligram mililiter milimole morpholiopropanesulfonic acid messenger RNA microgram microliter neomycin nucleotide(s) oligonucleotide ortho-nitrophenyl-β-galactoside optical Density at a wavelength of 578 nm phosphate-buffer saline gene coding for penicillin-binding protein PBP4* in B. subtilis an IPTG inducble promoter, a hybrid promoter of P groe and lac operator picromole an IPTG-inducible promoter, a hybrid promoter of the phage SPO-1 and the laco (glucose stavation inducible) σ B -dependent general stress promoter a glycine-inducible promoter, a promoter and 5 UTR of the gcvt operon a lysine-inducible promoter, a promoter and 5 UTR of the lysc operon ribosome binding site 138

147 Rif rifampicin RNA ribonucleic acid RNAP RNA polymerase rpm revolution or round per minute RT room temperature S. aureus Staphylococcus aureus sec second SD Shine-Dalgarno SDS sodium dodecyl sulphate srta gene coding for SrtA (e.i. of L. monocytogenes) TEMED N,N,N`,N`-tetramethylenethylendiamide TCA trichloroacetic acid Tris tri-(hydroxymethyl)-aminomethane Tween-20 polyoxyethylensorbitane monlaurate U units (enzyme activity) UTR untranslated region v/v volume/volume w/v weight/volume X-gal 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside yhcs gene coding for protein YhcS in B. subtilis ywpe gene coding for protein YwpE in B. subtilis 139

148 Erklärung Hiermit erkläre ich, dass ich die Arbeit selbständig verfasst und keine anderen als die von mir angegebenen Quellen und Hilfsmittel benutzt habe. Ferner erkläre ich, dass ich anderweitig mit oder ohne Erfolg nicht versucht habe, diese Dissertation einzureichen. Ich habe keine gleichartige Doktorprüfung an einer anderen Hochschule endgültig nicht bestanden. Bayreuth, März